wuggy/Dactyloidae
wuggy/Dactyloidae
1
0
Fork 0
mirror of https://repo.dactyloidae.xyz/Dactyloidae/UXP.git synced 2026-09-28 03:17:31 +09:00
Code Issues Projects Releases Packages Wiki Activity

update libaom to rev 1f0a4746aa8bfcce3fe7fbd2092aef8fe247aef4 (without moz.build and aom_ports/aom_once.h)

Browse source
This commit is contained in:
Roy Tam 2019-10-30 15:19:33 +08:00
commit 4f8655fb73
505 changed files with 50582 additions and 27164 deletions
Download patch file Download diff file Expand all files Collapse all files

2
media/libaom/README_MCP
View file

@ -10,4 +10,4 @@ The upstream aom git repository is:
https://aomedia.googlesource.com/aom
The git commit ID used was 76574b6c09515d6687ebfa9760319e521f5abeb3.
The git commit ID used was 1f0a4746aa8bfcce3fe7fbd2092aef8fe247aef4.

11
media/libaom/config/generic/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 0
ARCH_X86_64 equ 0
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 0
HAVE_AVX2 equ 0
HAVE_DSPR2 equ 0

11
media/libaom/config/generic/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 0
#define ARCH_X86_64 0
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 0
#define HAVE_AVX2 0
#define HAVE_DSPR2 0

18
media/libaom/config/generic/config/aom_dsp_rtcd.h
View file

@ -25,7 +25,7 @@ extern "C" {
void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
#define aom_blend_a64_hmask aom_blend_a64_hmask_c
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
#define aom_blend_a64_mask aom_blend_a64_mask_c
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -37,7 +37,7 @@ void aom_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *ds
void aom_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
#define aom_convolve8_vert aom_convolve8_vert_c
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
#define aom_convolve_copy aom_convolve_copy_c
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -340,25 +340,25 @@ void aom_h_predictor_8x4_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
#define aom_h_predictor_8x8 aom_h_predictor_8x8_c
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
#define aom_highbd_blend_a64_d16_mask aom_highbd_blend_a64_d16_mask_c
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
#define aom_highbd_blend_a64_hmask aom_highbd_blend_a64_hmask_c
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
#define aom_highbd_blend_a64_mask aom_highbd_blend_a64_mask_c
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
#define aom_highbd_blend_a64_vmask aom_highbd_blend_a64_vmask_c
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
#define aom_highbd_convolve8_horiz aom_highbd_convolve8_horiz_c
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
#define aom_highbd_convolve8_vert aom_highbd_convolve8_vert_c
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
#define aom_highbd_convolve_copy aom_highbd_convolve_copy_c
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -664,7 +664,7 @@ void aom_highbd_h_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
#define aom_highbd_lpf_horizontal_14 aom_highbd_lpf_horizontal_14_c
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
#define aom_highbd_lpf_horizontal_14_dual aom_highbd_lpf_horizontal_14_dual_c
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1009,7 +1009,7 @@ void aom_highbd_v_predictor_8x4_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
#define aom_highbd_v_predictor_8x8 aom_highbd_v_predictor_8x8_c
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
#define aom_lowbd_blend_a64_d16_mask aom_lowbd_blend_a64_d16_mask_c
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

128
media/libaom/config/generic/config/av1_rtcd.h
View file

@ -34,27 +34,50 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
#define apply_selfguided_restoration apply_selfguided_restoration_c
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
#define av1_apply_selfguided_restoration av1_apply_selfguided_restoration_c
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
#define av1_build_compound_diffwtd_mask av1_build_compound_diffwtd_mask_c
@ -65,34 +88,55 @@ void av1_build_compound_diffwtd_mask_d16_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask
void av1_build_compound_diffwtd_mask_highbd_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
#define av1_build_compound_diffwtd_mask_highbd av1_build_compound_diffwtd_mask_highbd_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
#define av1_calc_frame_error av1_calc_frame_error_c
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_convolve_2d_copy_sr av1_convolve_2d_copy_sr_c
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
#define av1_convolve_2d_scale av1_convolve_2d_scale_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_convolve_2d_sr av1_convolve_2d_sr_c
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
#define av1_convolve_horiz_rs av1_convolve_horiz_rs_c
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_convolve_x_sr av1_convolve_x_sr_c
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_convolve_y_sr av1_convolve_y_sr_c
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_dist_wtd_convolve_2d av1_dist_wtd_convolve_2d_c
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_dist_wtd_convolve_2d_copy av1_dist_wtd_convolve_2d_copy_c
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_dist_wtd_convolve_x av1_dist_wtd_convolve_x_c
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
#define av1_dist_wtd_convolve_y av1_dist_wtd_convolve_y_c
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -122,13 +166,13 @@ void av1_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint
void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_c
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_2d_copy_sr av1_highbd_convolve_2d_copy_sr_c
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_2d_scale av1_highbd_convolve_2d_scale_c
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_2d_sr av1_highbd_convolve_2d_sr_c
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
@ -140,22 +184,22 @@ void av1_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_
void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
#define av1_highbd_convolve_horiz_rs av1_highbd_convolve_horiz_rs_c
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_x_sr av1_highbd_convolve_x_sr_c
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_y_sr av1_highbd_convolve_y_sr_c
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_2d av1_highbd_dist_wtd_convolve_2d_c
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_2d_copy av1_highbd_dist_wtd_convolve_2d_copy_c
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_x av1_highbd_dist_wtd_convolve_x_c
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_y av1_highbd_dist_wtd_convolve_y_c
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -167,25 +211,25 @@ void av1_highbd_dr_prediction_z2_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
#define av1_highbd_dr_prediction_z3 av1_highbd_dr_prediction_z3_c
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add av1_highbd_inv_txfm_add_c
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_16x4 av1_highbd_inv_txfm_add_16x4_c
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_4x16 av1_highbd_inv_txfm_add_4x16_c
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_4x4 av1_highbd_inv_txfm_add_4x4_c
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_4x8 av1_highbd_inv_txfm_add_4x8_c
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_8x4 av1_highbd_inv_txfm_add_8x4_c
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_8x8 av1_highbd_inv_txfm_add_8x8_c
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
@ -197,7 +241,7 @@ void av1_highbd_iwht4x4_1_add_c(const tran_low_t *input, uint8_t *dest, int dest
void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
#define av1_highbd_warp_affine av1_highbd_warp_affine_c
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
#define av1_highbd_wiener_convolve_add_src av1_highbd_wiener_convolve_add_src_c
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
@ -280,6 +324,12 @@ void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int he
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
#define av1_wiener_convolve_add_src av1_wiener_convolve_add_src_c
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
#define cdef_copy_rect8_16bit_to_16bit cdef_copy_rect8_16bit_to_16bit_c
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
#define cdef_copy_rect8_8bit_to_16bit cdef_copy_rect8_8bit_to_16bit_c
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
#define cdef_filter_block cdef_filter_block_c
@ -304,20 +354,14 @@ cfl_subsample_hbd_fn cfl_get_luma_subsampling_444_hbd_c(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_c(TX_SIZE tx_size);
#define cfl_get_luma_subsampling_444_lbd cfl_get_luma_subsampling_444_lbd_c
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
#define copy_rect8_16bit_to_16bit copy_rect8_16bit_to_16bit_c
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
#define cfl_get_predict_hbd_fn cfl_get_predict_hbd_fn_c
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
#define copy_rect8_8bit_to_16bit copy_rect8_8bit_to_16bit_c
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
#define cfl_get_predict_lbd_fn cfl_get_predict_lbd_fn_c
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
#define get_predict_hbd_fn get_predict_hbd_fn_c
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
#define get_predict_lbd_fn get_predict_lbd_fn_c
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
#define get_subtract_average_fn get_subtract_average_fn_c
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
#define cfl_get_subtract_average_fn cfl_get_subtract_average_fn_c
void av1_rtcd(void);

11
media/libaom/config/linux/arm/config/aom_config.asm
View file

@ -14,12 +14,11 @@
.equ ARCH_PPC, 0
.equ ARCH_X86, 0
.equ ARCH_X86_64, 0
.equ CONFIG_2PASS_PARTITION_SEARCH_LVL_END, 3
.equ CONFIG_2PASS_PARTITION_SEARCH_LVL_START, 1
.equ CONFIG_ACCOUNTING, 0
.equ CONFIG_ANALYZER, 0
.equ CONFIG_AV1_DECODER, 1
.equ CONFIG_AV1_ENCODER, 0
.equ CONFIG_AV1_HIGHBITDEPTH, 1
.equ CONFIG_BIG_ENDIAN, 0
.equ CONFIG_BITSTREAM_DEBUG, 0
.equ CONFIG_COEFFICIENT_RANGE_CHECKING, 0
@ -31,22 +30,24 @@
.equ CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8, 1
.equ CONFIG_DIST_8X8, 0
.equ CONFIG_ENTROPY_STATS, 0
.equ CONFIG_FILEOPTIONS, 1
.equ CONFIG_GCC, 1
.equ CONFIG_GCOV, 0
.equ CONFIG_GPROF, 0
.equ CONFIG_HTB_TRELLIS, 0
.equ CONFIG_INSPECTION, 0
.equ CONFIG_INTERNAL_STATS, 0
.equ CONFIG_INTER_STATS_ONLY, 0
.equ CONFIG_LIBYUV, 0
.equ CONFIG_LOWBITDEPTH, 1
.equ CONFIG_LPF_MASK, 0
.equ CONFIG_MAX_DECODE_PROFILE, 2
.equ CONFIG_MISMATCH_DEBUG, 0
.equ CONFIG_MULTITHREAD, 1
.equ CONFIG_NN_V2, 0
.equ CONFIG_NORMAL_TILE_MODE, 0
.equ CONFIG_OS_SUPPORT, 1
.equ CONFIG_PIC, 1
.equ CONFIG_RD_DEBUG, 0
.equ CONFIG_REALTIME_ONLY, 0
.equ CONFIG_RUNTIME_CPU_DETECT, 1
.equ CONFIG_SHARED, 0
.equ CONFIG_SHARP_SETTINGS, 0
@ -54,9 +55,11 @@
.equ CONFIG_SPATIAL_RESAMPLING, 1
.equ CONFIG_SPEED_STATS, 0
.equ CONFIG_STATIC, 1
.equ CONFIG_SUPERRES_IN_RECODE, 1
.equ CONFIG_WEBM_IO, 0
.equ DECODE_HEIGHT_LIMIT, 0
.equ DECODE_WIDTH_LIMIT, 0
.equ FORCE_HIGHBITDEPTH_DECODING, 0
.equ HAVE_AVX, 0
.equ HAVE_AVX2, 0
.equ HAVE_DSPR2, 0

11
media/libaom/config/linux/arm/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 0
#define ARCH_X86_64 0
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 1
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 0
#define HAVE_AVX2 0
#define HAVE_DSPR2 0

22
media/libaom/config/linux/arm/config/aom_dsp_rtcd.h
View file

@ -26,7 +26,7 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_neon(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
#define aom_blend_a64_mask aom_blend_a64_mask_c
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -39,7 +39,7 @@ void aom_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *ds
void aom_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
#define aom_convolve8_vert aom_convolve8_vert_c
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
#define aom_convolve_copy aom_convolve_copy_c
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -362,25 +362,25 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_neon(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
RTCD_EXTERN void (*aom_h_predictor_8x8)(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
#define aom_highbd_blend_a64_d16_mask aom_highbd_blend_a64_d16_mask_c
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
#define aom_highbd_blend_a64_hmask aom_highbd_blend_a64_hmask_c
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
#define aom_highbd_blend_a64_mask aom_highbd_blend_a64_mask_c
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
#define aom_highbd_blend_a64_vmask aom_highbd_blend_a64_vmask_c
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
#define aom_highbd_convolve8_horiz aom_highbd_convolve8_horiz_c
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
#define aom_highbd_convolve8_vert aom_highbd_convolve8_vert_c
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
#define aom_highbd_convolve_copy aom_highbd_convolve_copy_c
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -691,7 +691,7 @@ void aom_highbd_h_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
#define aom_highbd_lpf_horizontal_14 aom_highbd_lpf_horizontal_14_c
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
#define aom_highbd_lpf_horizontal_14_dual aom_highbd_lpf_horizontal_14_dual_c
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1036,9 +1036,9 @@ void aom_highbd_v_predictor_8x4_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
#define aom_highbd_v_predictor_8x8 aom_highbd_v_predictor_8x8_c
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_neon(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_neon(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

207
media/libaom/config/linux/arm/config/av1_rtcd.h
View file

@ -34,28 +34,50 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_neon(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
#define av1_apply_selfguided_restoration av1_apply_selfguided_restoration_c
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
#define av1_build_compound_diffwtd_mask av1_build_compound_diffwtd_mask_c
@ -67,43 +89,64 @@ RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_d16)(uint8_t *mask, DIFFWTD_M
void av1_build_compound_diffwtd_mask_highbd_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
#define av1_build_compound_diffwtd_mask_highbd av1_build_compound_diffwtd_mask_highbd_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
#define av1_calc_frame_error av1_calc_frame_error_c
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
#define av1_convolve_2d_scale av1_convolve_2d_scale_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
#define av1_convolve_horiz_rs av1_convolve_horiz_rs_c
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_neon(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
#define av1_dr_prediction_z1 av1_dr_prediction_z1_c
@ -132,13 +175,13 @@ void av1_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint
void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_c
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_2d_copy_sr av1_highbd_convolve_2d_copy_sr_c
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_2d_scale av1_highbd_convolve_2d_scale_c
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_2d_sr av1_highbd_convolve_2d_sr_c
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
@ -150,22 +193,22 @@ void av1_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_
void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
#define av1_highbd_convolve_horiz_rs av1_highbd_convolve_horiz_rs_c
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_x_sr av1_highbd_convolve_x_sr_c
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_convolve_y_sr av1_highbd_convolve_y_sr_c
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_2d av1_highbd_dist_wtd_convolve_2d_c
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_2d_copy av1_highbd_dist_wtd_convolve_2d_copy_c
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_x av1_highbd_dist_wtd_convolve_x_c
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
#define av1_highbd_dist_wtd_convolve_y av1_highbd_dist_wtd_convolve_y_c
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -177,25 +220,25 @@ void av1_highbd_dr_prediction_z2_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
#define av1_highbd_dr_prediction_z3 av1_highbd_dr_prediction_z3_c
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add av1_highbd_inv_txfm_add_c
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_16x4 av1_highbd_inv_txfm_add_16x4_c
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_4x16 av1_highbd_inv_txfm_add_4x16_c
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_4x4 av1_highbd_inv_txfm_add_4x4_c
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_4x8 av1_highbd_inv_txfm_add_4x8_c
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_8x4 av1_highbd_inv_txfm_add_8x4_c
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
#define av1_highbd_inv_txfm_add_8x8 av1_highbd_inv_txfm_add_8x8_c
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
@ -207,7 +250,7 @@ void av1_highbd_iwht4x4_1_add_c(const tran_low_t *input, uint8_t *dest, int dest
void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
#define av1_highbd_warp_affine av1_highbd_warp_affine_c
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
#define av1_highbd_wiener_convolve_add_src av1_highbd_wiener_convolve_add_src_c
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
@ -268,7 +311,8 @@ void av1_inv_txfm2d_add_8x8_c(const int32_t *input, uint16_t *output, int stride
#define av1_inv_txfm2d_add_8x8 av1_inv_txfm2d_add_8x8_c
void av1_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
#define av1_inv_txfm_add av1_inv_txfm_add_c
void av1_inv_txfm_add_neon(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_round_shift_array_c(int32_t *arr, int size, int bit);
void av1_round_shift_array_neon(int32_t *arr, int size, int bit);
@ -277,12 +321,7 @@ RTCD_EXTERN void (*av1_round_shift_array)(int32_t *arr, int size, int bit);
int av1_selfguided_restoration_c(const uint8_t *dgd8, int width, int height,
int dgd_stride, int32_t *flt0, int32_t *flt1, int flt_stride,
int sgr_params_idx, int bit_depth, int highbd);
int av1_selfguided_restoration_neon(const uint8_t *dgd8, int width, int height,
int dgd_stride, int32_t *flt0, int32_t *flt1, int flt_stride,
int sgr_params_idx, int bit_depth, int highbd);
RTCD_EXTERN int (*av1_selfguided_restoration)(const uint8_t *dgd8, int width, int height,
int dgd_stride, int32_t *flt0, int32_t *flt1, int flt_stride,
int sgr_params_idx, int bit_depth, int highbd);
#define av1_selfguided_restoration av1_selfguided_restoration_c
void av1_upsample_intra_edge_c(uint8_t *p, int sz);
#define av1_upsample_intra_edge av1_upsample_intra_edge_c
@ -298,6 +337,14 @@ void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uin
void av1_wiener_convolve_add_src_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_neon(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_neon(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_neon(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
RTCD_EXTERN void (*cdef_filter_block)(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -330,25 +377,17 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_c(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_neon(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_neon(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_neon(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -362,8 +401,6 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_NEON) apply_selfguided_restoration = apply_selfguided_restoration_neon;
av1_build_compound_diffwtd_mask_d16 = av1_build_compound_diffwtd_mask_d16_c;
if (flags & HAS_NEON) av1_build_compound_diffwtd_mask_d16 = av1_build_compound_diffwtd_mask_d16_neon;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_c;
@ -382,14 +419,18 @@ static void setup_rtcd_internal(void)
if (flags & HAS_NEON) av1_dist_wtd_convolve_x = av1_dist_wtd_convolve_x_neon;
av1_dist_wtd_convolve_y = av1_dist_wtd_convolve_y_c;
if (flags & HAS_NEON) av1_dist_wtd_convolve_y = av1_dist_wtd_convolve_y_neon;
av1_inv_txfm_add = av1_inv_txfm_add_c;
if (flags & HAS_NEON) av1_inv_txfm_add = av1_inv_txfm_add_neon;
av1_round_shift_array = av1_round_shift_array_c;
if (flags & HAS_NEON) av1_round_shift_array = av1_round_shift_array_neon;
av1_selfguided_restoration = av1_selfguided_restoration_c;
if (flags & HAS_NEON) av1_selfguided_restoration = av1_selfguided_restoration_neon;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_NEON) av1_warp_affine = av1_warp_affine_neon;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_c;
if (flags & HAS_NEON) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_neon;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_c;
if (flags & HAS_NEON) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_neon;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_c;
if (flags & HAS_NEON) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_neon;
cdef_filter_block = cdef_filter_block_c;
if (flags & HAS_NEON) cdef_filter_block = cdef_filter_block_neon;
cdef_find_dir = cdef_find_dir_c;
@ -406,16 +447,12 @@ static void setup_rtcd_internal(void)
if (flags & HAS_NEON) cfl_get_luma_subsampling_444_hbd = cfl_get_luma_subsampling_444_hbd_neon;
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_NEON) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_neon;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_c;
if (flags & HAS_NEON) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_neon;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_c;
if (flags & HAS_NEON) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_neon;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_NEON) get_predict_hbd_fn = get_predict_hbd_fn_neon;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_NEON) get_predict_lbd_fn = get_predict_lbd_fn_neon;
get_subtract_average_fn = get_subtract_average_fn_c;
if (flags & HAS_NEON) get_subtract_average_fn = get_subtract_average_fn_neon;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_NEON) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_neon;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_NEON) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_neon;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_c;
if (flags & HAS_NEON) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_neon;
}
#endif

11
media/libaom/config/linux/ia32/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 1
ARCH_X86_64 equ 0
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 1
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/linux/ia32/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 1
#define ARCH_X86_64 0
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 1
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

68
media/libaom/config/linux/ia32/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,9 +47,9 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
RTCD_EXTERN void (*aom_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_dc_128_predictor_16x16_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
RTCD_EXTERN void (*aom_h_predictor_8x8)(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14)(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
RTCD_EXTERN void (*aom_highbd_v_predictor_8x8)(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

353
media/libaom/config/linux/ia32/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -155,19 +201,19 @@ void av1_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint
void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_c
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -179,35 +225,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -221,33 +267,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -259,10 +306,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -355,6 +402,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -362,6 +410,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -406,34 +468,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -445,9 +493,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -457,6 +505,9 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_c;
if (flags & HAS_SSE2) av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_c;
if (flags & HAS_SSE2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
@ -534,6 +585,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -569,9 +621,20 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_c;
if (flags & HAS_SSE2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_c;
if (flags & HAS_SSE2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_c;
if (flags & HAS_SSE2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_c;
if (flags & HAS_SSE2) cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
@ -600,25 +663,15 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_c;
if (flags & HAS_SSE2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_c;
if (flags & HAS_SSE2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_c;
if (flags & HAS_SSE2) get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_c;
if (flags & HAS_SSE2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

11
media/libaom/config/linux/x64/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 0
ARCH_X86_64 equ 1
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/linux/x64/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 0
#define ARCH_X86_64 1
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

66
media/libaom/config/linux/x64/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,8 +47,8 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
#define aom_convolve_copy aom_convolve_copy_sse2
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
#define aom_h_predictor_8x8 aom_h_predictor_8x8_sse2
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
#define aom_highbd_lpf_horizontal_14 aom_highbd_lpf_horizontal_14_sse2
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
#define aom_highbd_v_predictor_8x8 aom_highbd_v_predictor_8x8_sse2
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

346
media/libaom/config/linux/x64/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -158,19 +204,19 @@ void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8
void av1_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_sse2
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -182,35 +228,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -224,33 +270,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -262,10 +309,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -358,6 +405,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -365,6 +413,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -409,34 +471,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -448,9 +496,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -460,6 +508,8 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
av1_convolve_2d_scale = av1_convolve_2d_scale_c;
@ -528,6 +578,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -563,8 +614,17 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
if (flags & HAS_SSE4_1) cdef_filter_block = cdef_filter_block_sse4_1;
@ -591,22 +651,14 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

11
media/libaom/config/mac/x64/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 0
ARCH_X86_64 equ 1
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/mac/x64/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 0
#define ARCH_X86_64 1
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

66
media/libaom/config/mac/x64/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,8 +47,8 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
#define aom_convolve_copy aom_convolve_copy_sse2
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
#define aom_h_predictor_8x8 aom_h_predictor_8x8_sse2
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
#define aom_highbd_lpf_horizontal_14 aom_highbd_lpf_horizontal_14_sse2
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
#define aom_highbd_v_predictor_8x8 aom_highbd_v_predictor_8x8_sse2
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

346
media/libaom/config/mac/x64/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -158,19 +204,19 @@ void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8
void av1_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_sse2
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -182,35 +228,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -224,33 +270,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -262,10 +309,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -358,6 +405,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -365,6 +413,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -409,34 +471,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -448,9 +496,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -460,6 +508,8 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
av1_convolve_2d_scale = av1_convolve_2d_scale_c;
@ -528,6 +578,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -563,8 +614,17 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
if (flags & HAS_SSE4_1) cdef_filter_block = cdef_filter_block_sse4_1;
@ -591,22 +651,14 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

11
media/libaom/config/win/ia32/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 1
ARCH_X86_64 equ 0
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/win/ia32/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 1
#define ARCH_X86_64 0
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

68
media/libaom/config/win/ia32/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,9 +47,9 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
RTCD_EXTERN void (*aom_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_dc_128_predictor_16x16_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
RTCD_EXTERN void (*aom_h_predictor_8x8)(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14)(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
RTCD_EXTERN void (*aom_highbd_v_predictor_8x8)(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

353
media/libaom/config/win/ia32/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -155,19 +201,19 @@ void av1_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint
void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_c
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -179,35 +225,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -221,33 +267,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -259,10 +306,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -355,6 +402,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -362,6 +410,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -406,34 +468,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -445,9 +493,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -457,6 +505,9 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_c;
if (flags & HAS_SSE2) av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_c;
if (flags & HAS_SSE2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
@ -534,6 +585,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -569,9 +621,20 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_c;
if (flags & HAS_SSE2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_c;
if (flags & HAS_SSE2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_c;
if (flags & HAS_SSE2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_c;
if (flags & HAS_SSE2) cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
@ -600,25 +663,15 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_c;
if (flags & HAS_SSE2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_c;
if (flags & HAS_SSE2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_c;
if (flags & HAS_SSE2) get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_c;
if (flags & HAS_SSE2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

11
media/libaom/config/win/mingw32/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 1
ARCH_X86_64 equ 0
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/win/mingw32/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 1
#define ARCH_X86_64 0
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

68
media/libaom/config/win/mingw32/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,9 +47,9 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
RTCD_EXTERN void (*aom_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_dc_128_predictor_16x16_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
RTCD_EXTERN void (*aom_h_predictor_8x8)(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14)(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
RTCD_EXTERN void (*aom_highbd_v_predictor_8x8)(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

353
media/libaom/config/win/mingw32/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -155,19 +201,19 @@ void av1_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint
void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_c
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -179,35 +225,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -221,33 +267,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -259,10 +306,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -355,6 +402,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -362,6 +410,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -406,34 +468,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -445,9 +493,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -457,6 +505,9 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_c;
if (flags & HAS_SSE2) av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_c;
if (flags & HAS_SSE2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
@ -534,6 +585,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -569,9 +621,20 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_c;
if (flags & HAS_SSE2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_c;
if (flags & HAS_SSE2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_c;
if (flags & HAS_SSE2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_c;
if (flags & HAS_SSE2) cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
@ -600,25 +663,15 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_c;
if (flags & HAS_SSE2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_c;
if (flags & HAS_SSE2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_c;
if (flags & HAS_SSE2) get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_c;
if (flags & HAS_SSE2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

11
media/libaom/config/win/mingw64/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 0
ARCH_X86_64 equ 1
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/win/mingw64/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 0
#define ARCH_X86_64 1
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

66
media/libaom/config/win/mingw64/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,8 +47,8 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
#define aom_convolve_copy aom_convolve_copy_sse2
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
#define aom_h_predictor_8x8 aom_h_predictor_8x8_sse2
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
#define aom_highbd_lpf_horizontal_14 aom_highbd_lpf_horizontal_14_sse2
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
#define aom_highbd_v_predictor_8x8 aom_highbd_v_predictor_8x8_sse2
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

346
media/libaom/config/win/mingw64/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -158,19 +204,19 @@ void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8
void av1_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_sse2
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -182,35 +228,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -224,33 +270,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -262,10 +309,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -358,6 +405,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -365,6 +413,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -409,34 +471,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -448,9 +496,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -460,6 +508,8 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
av1_convolve_2d_scale = av1_convolve_2d_scale_c;
@ -528,6 +578,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -563,8 +614,17 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
if (flags & HAS_SSE4_1) cdef_filter_block = cdef_filter_block_sse4_1;
@ -591,22 +651,14 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

11
media/libaom/config/win/x64/config/aom_config.asm
View file

@ -14,12 +14,11 @@ ARCH_MIPS equ 0
ARCH_PPC equ 0
ARCH_X86 equ 0
ARCH_X86_64 equ 1
CONFIG_2PASS_PARTITION_SEARCH_LVL_END equ 3
CONFIG_2PASS_PARTITION_SEARCH_LVL_START equ 1
CONFIG_ACCOUNTING equ 0
CONFIG_ANALYZER equ 0
CONFIG_AV1_DECODER equ 1
CONFIG_AV1_ENCODER equ 0
CONFIG_AV1_HIGHBITDEPTH equ 1
CONFIG_BIG_ENDIAN equ 0
CONFIG_BITSTREAM_DEBUG equ 0
CONFIG_COEFFICIENT_RANGE_CHECKING equ 0
@ -31,22 +30,24 @@ CONFIG_DENOISE equ 1
CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 equ 1
CONFIG_DIST_8X8 equ 0
CONFIG_ENTROPY_STATS equ 0
CONFIG_FILEOPTIONS equ 1
CONFIG_GCC equ 1
CONFIG_GCOV equ 0
CONFIG_GPROF equ 0
CONFIG_HTB_TRELLIS equ 0
CONFIG_INSPECTION equ 0
CONFIG_INTERNAL_STATS equ 0
CONFIG_INTER_STATS_ONLY equ 0
CONFIG_LIBYUV equ 0
CONFIG_LOWBITDEPTH equ 1
CONFIG_LPF_MASK equ 0
CONFIG_MAX_DECODE_PROFILE equ 2
CONFIG_MISMATCH_DEBUG equ 0
CONFIG_MULTITHREAD equ 1
CONFIG_NN_V2 equ 0
CONFIG_NORMAL_TILE_MODE equ 0
CONFIG_OS_SUPPORT equ 1
CONFIG_PIC equ 0
CONFIG_RD_DEBUG equ 0
CONFIG_REALTIME_ONLY equ 0
CONFIG_RUNTIME_CPU_DETECT equ 1
CONFIG_SHARED equ 0
CONFIG_SHARP_SETTINGS equ 0
@ -54,9 +55,11 @@ CONFIG_SIZE_LIMIT equ 0
CONFIG_SPATIAL_RESAMPLING equ 1
CONFIG_SPEED_STATS equ 0
CONFIG_STATIC equ 1
CONFIG_SUPERRES_IN_RECODE equ 1
CONFIG_WEBM_IO equ 0
DECODE_HEIGHT_LIMIT equ 0
DECODE_WIDTH_LIMIT equ 0
FORCE_HIGHBITDEPTH_DECODING equ 0
HAVE_AVX equ 1
HAVE_AVX2 equ 1
HAVE_DSPR2 equ 0

11
media/libaom/config/win/x64/config/aom_config.h
View file

@ -16,12 +16,11 @@
#define ARCH_PPC 0
#define ARCH_X86 0
#define ARCH_X86_64 1
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_END 3
#define CONFIG_2PASS_PARTITION_SEARCH_LVL_START 1
#define CONFIG_ACCOUNTING 0
#define CONFIG_ANALYZER 0
#define CONFIG_AV1_DECODER 1
#define CONFIG_AV1_ENCODER 0
#define CONFIG_AV1_HIGHBITDEPTH 1
#define CONFIG_BIG_ENDIAN 0
#define CONFIG_BITSTREAM_DEBUG 0
#define CONFIG_COEFFICIENT_RANGE_CHECKING 0
@ -33,22 +32,24 @@
#define CONFIG_DISABLE_FULL_PIXEL_SPLIT_8X8 1
#define CONFIG_DIST_8X8 0
#define CONFIG_ENTROPY_STATS 0
#define CONFIG_FILEOPTIONS 1
#define CONFIG_GCC 1
#define CONFIG_GCOV 0
#define CONFIG_GPROF 0
#define CONFIG_HTB_TRELLIS 0
#define CONFIG_INSPECTION 0
#define CONFIG_INTERNAL_STATS 0
#define CONFIG_INTER_STATS_ONLY 0
#define CONFIG_LIBYUV 0
#define CONFIG_LOWBITDEPTH 1
#define CONFIG_LPF_MASK 0
#define CONFIG_MAX_DECODE_PROFILE 2
#define CONFIG_MISMATCH_DEBUG 0
#define CONFIG_MULTITHREAD 1
#define CONFIG_NN_V2 0
#define CONFIG_NORMAL_TILE_MODE 0
#define CONFIG_OS_SUPPORT 1
#define CONFIG_PIC 0
#define CONFIG_RD_DEBUG 0
#define CONFIG_REALTIME_ONLY 0
#define CONFIG_RUNTIME_CPU_DETECT 1
#define CONFIG_SHARED 0
#define CONFIG_SHARP_SETTINGS 0
@ -56,9 +57,11 @@
#define CONFIG_SPATIAL_RESAMPLING 1
#define CONFIG_SPEED_STATS 0
#define CONFIG_STATIC 1
#define CONFIG_SUPERRES_IN_RECODE 1
#define CONFIG_WEBM_IO 0
#define DECODE_HEIGHT_LIMIT 0
#define DECODE_WIDTH_LIMIT 0
#define FORCE_HIGHBITDEPTH_DECODING 0
#define HAVE_AVX 1
#define HAVE_AVX2 1
#define HAVE_DSPR2 0

66
media/libaom/config/win/x64/config/aom_dsp_rtcd.h
View file

@ -26,10 +26,10 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src
void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
RTCD_EXTERN void (*aom_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby);
void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
RTCD_EXTERN void (*aom_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh);
void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h);
@ -47,8 +47,8 @@ void aom_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t
void aom_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
RTCD_EXTERN void (*aom_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h);
void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
void aom_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h);
#define aom_convolve_copy aom_convolve_copy_sse2
void aom_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
@ -471,37 +471,37 @@ void aom_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *abov
void aom_h_predictor_8x8_sse2(uint8_t *dst, ptrdiff_t y_stride, const uint8_t *above, const uint8_t *left);
#define aom_h_predictor_8x8 aom_h_predictor_8x8_sse2
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params, const int bd);
void aom_highbd_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_hmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, int bd);
void aom_highbd_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_mask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, int bd);
void aom_highbd_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_blend_a64_vmask)(uint8_t *dst, uint32_t dst_stride, const uint8_t *src0, uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride, const uint8_t *mask, int w, int h, int bd);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bd);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
void aom_highbd_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_convolve_copy_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
RTCD_EXTERN void (*aom_highbd_convolve_copy)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int filter_x_stride, const int16_t *filter_y, int filter_y_stride, int w, int h, int bd);
void aom_highbd_dc_128_predictor_16x16_c(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
void aom_highbd_dc_128_predictor_16x16_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
@ -857,10 +857,10 @@ void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch, const uint8_t *blimi
void aom_highbd_lpf_horizontal_14_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
#define aom_highbd_lpf_horizontal_14 aom_highbd_lpf_horizontal_14_sse2
void aom_highbd_lpf_horizontal_14_dual_c(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limt1, const uint8_t *thresh1,int bd);
void aom_highbd_lpf_horizontal_14_dual_c(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_sse2(uint16_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,int bd);
void aom_highbd_lpf_horizontal_14_dual_avx2(uint16_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,int bd);
RTCD_EXTERN void (*aom_highbd_lpf_horizontal_14_dual)(uint16_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,int bd);
void aom_highbd_lpf_horizontal_4_c(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
void aom_highbd_lpf_horizontal_4_sse2(uint16_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int bd);
@ -1233,10 +1233,10 @@ void aom_highbd_v_predictor_8x8_c(uint16_t *dst, ptrdiff_t y_stride, const uint1
void aom_highbd_v_predictor_8x8_sse2(uint16_t *dst, ptrdiff_t y_stride, const uint16_t *above, const uint16_t *left, int bd);
#define aom_highbd_v_predictor_8x8 aom_highbd_v_predictor_8x8_sse2
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subx, int suby, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_c(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_sse4_1(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lowbd_blend_a64_d16_mask_avx2(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
RTCD_EXTERN void (*aom_lowbd_blend_a64_d16_mask)(uint8_t *dst, uint32_t dst_stride, const CONV_BUF_TYPE *src0, uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride, const uint8_t *mask, uint32_t mask_stride, int w, int h, int subw, int subh, ConvolveParams *conv_params);
void aom_lpf_horizontal_14_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);
void aom_lpf_horizontal_14_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh);

346
media/libaom/config/win/x64/config/av1_rtcd.h
View file

@ -34,29 +34,52 @@ struct yv12_buffer_config;
struct NN_CONFIG;
typedef struct NN_CONFIG NN_CONFIG;
enum { NONE, RELU, SOFTSIGN, SIGMOID } UENUM1BYTE(ACTIVATION);
#if CONFIG_NN_V2
enum { SOFTMAX_CROSS_ENTROPY } UENUM1BYTE(LOSS);
struct NN_CONFIG_V2;
typedef struct NN_CONFIG_V2 NN_CONFIG_V2;
struct FC_LAYER;
typedef struct FC_LAYER FC_LAYER;
#endif // CONFIG_NN_V2
struct CNN_CONFIG;
typedef struct CNN_CONFIG CNN_CONFIG;
struct CNN_LAYER_CONFIG;
typedef struct CNN_LAYER_CONFIG CNN_LAYER_CONFIG;
struct CNN_THREAD_DATA;
typedef struct CNN_THREAD_DATA CNN_THREAD_DATA;
struct CNN_BRANCH_CONFIG;
typedef struct CNN_BRANCH_CONFIG CNN_BRANCH_CONFIG;
struct CNN_MULTI_OUT;
typedef struct CNN_MULTI_OUT CNN_MULTI_OUT;
/* Function pointers return by CfL functions */
typedef void (*cfl_subsample_lbd_fn)(const uint8_t *input, int input_stride,
uint16_t *output_q3);
#if CONFIG_AV1_HIGHBITDEPTH
typedef void (*cfl_subsample_hbd_fn)(const uint16_t *input, int input_stride,
uint16_t *output_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#endif
typedef void (*cfl_subtract_average_fn)(const uint16_t *src, int16_t *dst);
typedef void (*cfl_predict_lbd_fn)(const int16_t *src, uint8_t *dst,
int dst_stride, int alpha_q3);
typedef void (*cfl_predict_hbd_fn)(const int16_t *src, uint16_t *dst,
int dst_stride, int alpha_q3, int bd);
#ifdef __cplusplus
extern "C" {
#endif
void apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_c(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_sse4_1(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_apply_selfguided_restoration_avx2(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
RTCD_EXTERN void (*av1_apply_selfguided_restoration)(const uint8_t *dat, int width, int height, int stride, int eps, const int *xqd, uint8_t *dst, int dst_stride, int32_t *tmpbuf, int bit_depth, int highbd);
void av1_build_compound_diffwtd_mask_c(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
void av1_build_compound_diffwtd_mask_sse4_1(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w);
@ -73,54 +96,77 @@ void av1_build_compound_diffwtd_mask_highbd_ssse3(uint8_t *mask, DIFFWTD_MASK_TY
void av1_build_compound_diffwtd_mask_highbd_avx2(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
RTCD_EXTERN void (*av1_build_compound_diffwtd_mask_highbd)(uint8_t *mask, DIFFWTD_MASK_TYPE mask_type, const uint8_t *src0, int src0_stride, const uint8_t *src1, int src1_stride, int h, int w, int bd);
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
int64_t av1_calc_frame_error_c(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_sse2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
int64_t av1_calc_frame_error_avx2(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
RTCD_EXTERN int64_t (*av1_calc_frame_error)(const uint8_t *const ref, int stride, const uint8_t *const dst, int p_width, int p_height, int p_stride);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params);
void av1_cnn_activate_c( float **input, int channels, int width, int height, int stride, ACTIVATION layer_activation);
#define av1_cnn_activate av1_cnn_activate_c
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_cnn_add_c( float **input, int channels, int width, int height, int stride, const float **add);
#define av1_cnn_add av1_cnn_add_c
void av1_cnn_batchnorm_c(float **image, int channels, int width, int height, int stride, const float *gamma, const float *beta, const float *mean, const float *std);
#define av1_cnn_batchnorm av1_cnn_batchnorm_c
void av1_cnn_convolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride, int start_idx, int step);
#define av1_cnn_convolve av1_cnn_convolve_c
void av1_cnn_deconvolve_c( const float **input, int in_width, int in_height, int in_stride, const CNN_LAYER_CONFIG *layer_config, float **output, int out_stride);
#define av1_cnn_deconvolve av1_cnn_deconvolve_c
void av1_cnn_predict_c( const float **input, int in_width, int in_height, int in_stride, const CNN_CONFIG *cnn_config, const CNN_THREAD_DATA *thread_data, CNN_MULTI_OUT *output_struct);
#define av1_cnn_predict av1_cnn_predict_c
void av1_convolve_2d_copy_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_copy_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_copy_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_scale_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_scale)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_2d_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_2d_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_horiz_rs_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_horiz_rs_sse4_1(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
RTCD_EXTERN void (*av1_convolve_horiz_rs)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_x_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_x_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_x_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_convolve_y_sr_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_convolve_y_sr_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_convolve_y_sr)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_2d_copy_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_2d_copy)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_x_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_x)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_sse2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dist_wtd_convolve_y_avx2(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_dist_wtd_convolve_y)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params);
void av1_dr_prediction_z1_c(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
void av1_dr_prediction_z1_avx2(uint8_t *dst, ptrdiff_t stride, int bw, int bh, const uint8_t *above, const uint8_t *left, int upsample_above, int dx, int dy);
@ -158,19 +204,19 @@ void av1_highbd_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8
void av1_highbd_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_sse2
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_sse2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_copy_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_copy_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int x_step_qn, const int subpel_y_q4, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_scale_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_scale)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int x_step_qn, const int subpel_y_qn, const int y_step_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_2d_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_2d_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, int bps);
#define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
@ -182,35 +228,35 @@ void av1_highbd_convolve_horiz_rs_c(const uint16_t *src, int src_stride, uint16_
void av1_highbd_convolve_horiz_rs_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_horiz_rs)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const int16_t *x_filters, int x0_qn, int x_step_qn, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_x_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_convolve_y_sr)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_2d_copy_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_2d_copy)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_x_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_x)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dist_wtd_convolve_y_avx2(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_dist_wtd_convolve_y)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams *filter_params_x, const InterpFilterParams *filter_params_y, const int subpel_x_qn, const int subpel_y_qn, ConvolveParams *conv_params, int bd);
void av1_highbd_dr_prediction_z1_c(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
void av1_highbd_dr_prediction_z1_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_above, int dx, int dy, int bd);
@ -224,33 +270,34 @@ void av1_highbd_dr_prediction_z3_c(uint16_t *dst, ptrdiff_t stride, int bw, int
void av1_highbd_dr_prediction_z3_avx2(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
RTCD_EXTERN void (*av1_highbd_dr_prediction_z3)(uint16_t *dst, ptrdiff_t stride, int bw, int bh, const uint16_t *above, const uint16_t *left, int upsample_left, int dx, int dy, int bd);
void av1_highbd_inv_txfm_add_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_avx2(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_16x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_16x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x16_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x16)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_4x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_4x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x4_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x4)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *dqcoeff, uint8_t *dst, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_c(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_inv_txfm_add_8x8_sse4_1(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
RTCD_EXTERN void (*av1_highbd_inv_txfm_add_8x8)(const tran_low_t *input, uint8_t *dest, int stride, const TxfmParam *txfm_param);
void av1_highbd_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int bd);
#define av1_highbd_iwht4x4_16_add av1_highbd_iwht4x4_16_add_c
@ -262,10 +309,10 @@ void av1_highbd_warp_affine_c(const int32_t *mat, const uint16_t *ref, int width
void av1_highbd_warp_affine_sse4_1(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_highbd_warp_affine)(const int32_t *mat, const uint16_t *ref, int width, int height, int stride, uint16_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, int bd, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bps);
void av1_highbd_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_highbd_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
RTCD_EXTERN void (*av1_highbd_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params, int bd);
void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, TX_TYPE tx_type, int bd);
#define av1_inv_txfm2d_add_16x16 av1_inv_txfm2d_add_16x16_c
@ -358,6 +405,7 @@ RTCD_EXTERN void (*av1_upsample_intra_edge_high)(uint16_t *p, int sz, int bd);
void av1_warp_affine_c(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_sse4_1(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_warp_affine_avx2(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
RTCD_EXTERN void (*av1_warp_affine)(const int32_t *mat, const uint8_t *ref, int width, int height, int stride, uint8_t *pred, int p_col, int p_row, int p_width, int p_height, int p_stride, int subsampling_x, int subsampling_y, ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
void av1_wiener_convolve_add_src_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
@ -365,6 +413,20 @@ void av1_wiener_convolve_add_src_sse2(const uint8_t *src, ptrdiff_t src_stride,
void av1_wiener_convolve_add_src_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
RTCD_EXTERN void (*av1_wiener_convolve_add_src)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h, const ConvolveParams *conv_params);
void cdef_copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*cdef_copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void cdef_filter_block_c(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_sse2(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
void cdef_filter_block_ssse3(uint8_t *dst8, uint16_t *dst16, int dstride, const uint16_t *in, int pri_strength, int sec_strength, int dir, int pri_damping, int sec_damping, int bsize, int coeff_shift);
@ -409,34 +471,20 @@ cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_ssse3(TX_SIZE tx_size);
cfl_subsample_lbd_fn cfl_get_luma_subsampling_444_lbd_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subsample_lbd_fn (*cfl_get_luma_subsampling_444_lbd)(TX_SIZE tx_size);
void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
void copy_rect8_16bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn cfl_get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*cfl_get_predict_hbd_fn)(TX_SIZE tx_size);
void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
void copy_rect8_8bit_to_16bit_avx2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn cfl_get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*cfl_get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_c(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_hbd_fn get_predict_hbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_hbd_fn (*get_predict_hbd_fn)(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_c(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_ssse3(TX_SIZE tx_size);
cfl_predict_lbd_fn get_predict_lbd_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_predict_lbd_fn (*get_predict_lbd_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*get_subtract_average_fn)(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_c(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_sse2(TX_SIZE tx_size);
cfl_subtract_average_fn cfl_get_subtract_average_fn_avx2(TX_SIZE tx_size);
RTCD_EXTERN cfl_subtract_average_fn (*cfl_get_subtract_average_fn)(TX_SIZE tx_size);
void av1_rtcd(void);
@ -448,9 +496,9 @@ static void setup_rtcd_internal(void)
(void)flags;
apply_selfguided_restoration = apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) apply_selfguided_restoration = apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) apply_selfguided_restoration = apply_selfguided_restoration_avx2;
av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_c;
if (flags & HAS_SSE4_1) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_sse4_1;
if (flags & HAS_AVX2) av1_apply_selfguided_restoration = av1_apply_selfguided_restoration_avx2;
av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_c;
if (flags & HAS_SSE4_1) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_sse4_1;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask = av1_build_compound_diffwtd_mask_avx2;
@ -460,6 +508,8 @@ static void setup_rtcd_internal(void)
av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_c;
if (flags & HAS_SSSE3) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_ssse3;
if (flags & HAS_AVX2) av1_build_compound_diffwtd_mask_highbd = av1_build_compound_diffwtd_mask_highbd_avx2;
av1_calc_frame_error = av1_calc_frame_error_sse2;
if (flags & HAS_AVX2) av1_calc_frame_error = av1_calc_frame_error_avx2;
av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_sse2;
if (flags & HAS_AVX2) av1_convolve_2d_copy_sr = av1_convolve_2d_copy_sr_avx2;
av1_convolve_2d_scale = av1_convolve_2d_scale_c;
@ -528,6 +578,7 @@ static void setup_rtcd_internal(void)
if (flags & HAS_AVX2) av1_highbd_dr_prediction_z3 = av1_highbd_dr_prediction_z3_avx2;
av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_sse4_1;
if (flags & HAS_AVX2) av1_highbd_inv_txfm_add = av1_highbd_inv_txfm_add_avx2;
av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_c;
if (flags & HAS_SSE4_1) av1_highbd_inv_txfm_add_16x4 = av1_highbd_inv_txfm_add_16x4_sse4_1;
av1_highbd_inv_txfm_add_4x16 = av1_highbd_inv_txfm_add_4x16_c;
@ -563,8 +614,17 @@ static void setup_rtcd_internal(void)
if (flags & HAS_SSE4_1) av1_upsample_intra_edge_high = av1_upsample_intra_edge_high_sse4_1;
av1_warp_affine = av1_warp_affine_c;
if (flags & HAS_SSE4_1) av1_warp_affine = av1_warp_affine_sse4_1;
if (flags & HAS_AVX2) av1_warp_affine = av1_warp_affine_avx2;
av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_sse2;
if (flags & HAS_AVX2) av1_wiener_convolve_add_src = av1_wiener_convolve_add_src_avx2;
cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_16bit_to_16bit = cdef_copy_rect8_16bit_to_16bit_avx2;
cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) cdef_copy_rect8_8bit_to_16bit = cdef_copy_rect8_8bit_to_16bit_avx2;
cdef_filter_block = cdef_filter_block_sse2;
if (flags & HAS_SSSE3) cdef_filter_block = cdef_filter_block_ssse3;
if (flags & HAS_SSE4_1) cdef_filter_block = cdef_filter_block_sse4_1;
@ -591,22 +651,14 @@ static void setup_rtcd_internal(void)
cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_c;
if (flags & HAS_SSSE3) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_ssse3;
if (flags & HAS_AVX2) cfl_get_luma_subsampling_444_lbd = cfl_get_luma_subsampling_444_lbd_avx2;
copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_avx2;
copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
if (flags & HAS_AVX2) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_avx2;
get_predict_hbd_fn = get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) get_predict_hbd_fn = get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_hbd_fn = get_predict_hbd_fn_avx2;
get_predict_lbd_fn = get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) get_predict_lbd_fn = get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) get_predict_lbd_fn = get_predict_lbd_fn_avx2;
get_subtract_average_fn = get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) get_subtract_average_fn = get_subtract_average_fn_avx2;
cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_hbd_fn = cfl_get_predict_hbd_fn_avx2;
cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_c;
if (flags & HAS_SSSE3) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_ssse3;
if (flags & HAS_AVX2) cfl_get_predict_lbd_fn = cfl_get_predict_lbd_fn_avx2;
cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_sse2;
if (flags & HAS_AVX2) cfl_get_subtract_average_fn = cfl_get_subtract_average_fn_avx2;
}
#endif

7
media/libaom/sources.mozbuild
View file

@ -111,6 +111,7 @@ files = {
'../../third_party/aom/av1/decoder/decodetxb.c',
'../../third_party/aom/av1/decoder/detokenize.c',
'../../third_party/aom/av1/decoder/obu.c',
'../../third_party/aom/av1/encoder/arm/neon/av1_error_neon.c',
'../../third_party/aom/av1/encoder/arm/neon/quantize_neon.c',
],
'GENERIC_EXPORTS': [
@ -330,7 +331,6 @@ files = {
'../../third_party/aom/av1/common/warped_motion.c',
'../../third_party/aom/av1/common/x86/av1_convolve_horiz_rs_sse4.c',
'../../third_party/aom/av1/common/x86/av1_convolve_scale_sse4.c',
'../../third_party/aom/av1/common/x86/av1_highbd_convolve_sse4.c',
'../../third_party/aom/av1/common/x86/av1_inv_txfm_avx2.c',
'../../third_party/aom/av1/common/x86/av1_inv_txfm_ssse3.c',
'../../third_party/aom/av1/common/x86/av1_txfm_sse4.c',
@ -362,6 +362,8 @@ files = {
'../../third_party/aom/av1/common/x86/reconinter_ssse3.c',
'../../third_party/aom/av1/common/x86/selfguided_avx2.c',
'../../third_party/aom/av1/common/x86/selfguided_sse4.c',
'../../third_party/aom/av1/common/x86/warp_plane_avx2.c',
'../../third_party/aom/av1/common/x86/warp_plane_sse2.c',
'../../third_party/aom/av1/common/x86/warp_plane_sse4.c',
'../../third_party/aom/av1/common/x86/wiener_convolve_avx2.c',
'../../third_party/aom/av1/common/x86/wiener_convolve_sse2.c',
@ -496,7 +498,6 @@ files = {
'../../third_party/aom/av1/common/warped_motion.c',
'../../third_party/aom/av1/common/x86/av1_convolve_horiz_rs_sse4.c',
'../../third_party/aom/av1/common/x86/av1_convolve_scale_sse4.c',
'../../third_party/aom/av1/common/x86/av1_highbd_convolve_sse4.c',
'../../third_party/aom/av1/common/x86/av1_inv_txfm_avx2.c',
'../../third_party/aom/av1/common/x86/av1_inv_txfm_ssse3.c',
'../../third_party/aom/av1/common/x86/av1_txfm_sse4.c',
@ -528,6 +529,8 @@ files = {
'../../third_party/aom/av1/common/x86/reconinter_ssse3.c',
'../../third_party/aom/av1/common/x86/selfguided_avx2.c',
'../../third_party/aom/av1/common/x86/selfguided_sse4.c',
'../../third_party/aom/av1/common/x86/warp_plane_avx2.c',
'../../third_party/aom/av1/common/x86/warp_plane_sse2.c',
'../../third_party/aom/av1/common/x86/warp_plane_sse4.c',
'../../third_party/aom/av1/common/x86/wiener_convolve_avx2.c',
'../../third_party/aom/av1/common/x86/wiener_convolve_sse2.c',

66
third_party/aom/.cmake-format.py vendored
View file

@ -1,13 +1,11 @@
# Generated with cmake-format 0.3.6
# Generated with cmake-format 0.5.1
# How wide to allow formatted cmake files
line_width = 80
# How many spaces to tab for indent
tab_size = 2
# If arglists are longer than this, break them always. This introduces some
# interesting effects with complicated 'if' statements. However, we want file
# lists to look reasonable. Try to strike a balance.
# If arglists are longer than this, break them always
max_subargs_per_line = 10
# If true, separate flow control names from their parentheses with a space
@ -21,10 +19,10 @@ separate_fn_name_with_space = False
dangle_parens = False
# What character to use for bulleted lists
bullet_char = u'*'
bullet_char = '*'
# What character to use as punctuation after numerals in an enumerated list
enum_char = u'.'
enum_char = '.'
# What style line endings to use in the output.
line_ending = u'unix'
@ -32,6 +30,9 @@ line_ending = u'unix'
# Format command names consistently as 'lower' or 'upper' case
command_case = u'lower'
# Format keywords consistently as 'lower' or 'upper' case
keyword_case = u'unchanged'
# Specify structure for custom cmake functions
additional_commands = {
"foo": {
@ -46,3 +47,56 @@ additional_commands = {
}
}
}
# A list of command names which should always be wrapped
always_wrap = []
# Specify the order of wrapping algorithms during successive reflow attempts
algorithm_order = [0, 1, 2, 3, 4]
# If true, the argument lists which are known to be sortable will be sorted
# lexicographicall
autosort = False
# enable comment markup parsing and reflow
enable_markup = True
# If comment markup is enabled, don't reflow the first comment block in
# eachlistfile. Use this to preserve formatting of your
# copyright/licensestatements.
first_comment_is_literal = False
# If comment markup is enabled, don't reflow any comment block which matchesthis
# (regex) pattern. Default is `None` (disabled).
literal_comment_pattern = None
# Regular expression to match preformat fences in comments
# default=r'^\s*([`~]{3}[`~]*)(.*)$'
fence_pattern = u'^\\s*([`~]{3}[`~]*)(.*)$'
# Regular expression to match rulers in comments
# default=r'^\s*[^\w\s]{3}.*[^\w\s]{3}$'
ruler_pattern = u'^\\s*[^\\w\\s]{3}.*[^\\w\\s]{3}$'
# If true, emit the unicode byte-order mark (BOM) at the start of the file
emit_byteorder_mark = False
# If a comment line starts with at least this many consecutive hash characters,
# then don't lstrip() them off. This allows for lazy hash rulers where the first
# hash char is not separated by space
hashruler_min_length = 10
# If true, then insert a space between the first hash char and remaining hash
# chars in a hash ruler, and normalize it's length to fill the column
canonicalize_hashrulers = True
# Specify the encoding of the input file. Defaults to utf-8.
input_encoding = u'utf-8'
# Specify the encoding of the output file. Defaults to utf-8. Note that cmake
# only claims to support utf-8 so be careful when using anything else
output_encoding = u'utf-8'
# A dictionary containing any per-command configuration overrides. Currently
# only `command_case` is supported.
per_command = {}

148
third_party/aom/CMakeLists.txt vendored
View file

@ -9,26 +9,27 @@
# can obtain it at www.aomedia.org/license/patent.
#
cmake_minimum_required(VERSION 3.5)
project(AOM C CXX)
if(NOT EMSCRIPTEN)
if(NOT CMAKE_BUILD_TYPE)
set(CMAKE_BUILD_TYPE "Release"
CACHE "Build type: Debug, Release, RelWithDebInfo or MinSizeRel" STRING
if(NOT CMAKE_BUILD_TYPE AND NOT CMAKE_CONFIGURATION_TYPES)
set(CMAKE_BUILD_TYPE
"Release"
CACHE STRING "Build type: Debug, Release, RelWithDebInfo or MinSizeRel"
FORCE)
endif()
endif()
project(AOM C CXX)
set(AOM_ROOT "${CMAKE_CURRENT_SOURCE_DIR}")
set(AOM_CONFIG_DIR "${CMAKE_CURRENT_BINARY_DIR}")
if("${AOM_ROOT}" STREQUAL "${AOM_CONFIG_DIR}")
message(FATAL_ERROR
"Building from within the aom source tree is not supported.\n"
"Hint: Run these commands\n" "$ rm -rf CMakeCache.txt CMakeFiles\n"
"$ mkdir -p ../aom_build\n" "$ cd ../aom_build\n"
"And re-run CMake from the aom_build directory.")
message(
FATAL_ERROR "Building from within the aom source tree is not supported.\n"
"Hint: Run these commands\n"
"$ rm -rf CMakeCache.txt CMakeFiles\n"
"$ mkdir -p ../aom_build\n" "$ cd ../aom_build\n"
"And re-run CMake from the aom_build directory.")
endif()
include("${AOM_ROOT}/build/cmake/aom_configure.cmake")
@ -113,6 +114,7 @@ list(APPEND AOM_SOURCES
"${AOM_ROOT}/aom/aomcx.h"
"${AOM_ROOT}/aom/aomdx.h"
"${AOM_ROOT}/aom/internal/aom_codec_internal.h"
"${AOM_ROOT}/aom/internal/aom_image_internal.h"
"${AOM_ROOT}/aom/src/aom_codec.c"
"${AOM_ROOT}/aom/src/aom_decoder.c"
"${AOM_ROOT}/aom/src/aom_encoder.c"
@ -176,15 +178,18 @@ add_custom_command(OUTPUT "${AOM_CONFIG_DIR}/config/aom_version.h"
-DGIT_EXECUTABLE=${GIT_EXECUTABLE}
-DPERL_EXECUTABLE=${PERL_EXECUTABLE} -P
"${AOM_ROOT}/build/cmake/version.cmake"
COMMENT "Writing aom_version.h" VERBATIM)
COMMENT "Writing aom_version.h"
VERBATIM)
add_custom_target(aom_version_check
COMMAND ${CMAKE_COMMAND} -DAOM_CONFIG_DIR=${AOM_CONFIG_DIR}
COMMAND ${CMAKE_COMMAND}
-DAOM_CONFIG_DIR=${AOM_CONFIG_DIR}
-DAOM_ROOT=${AOM_ROOT}
-DGIT_EXECUTABLE=${GIT_EXECUTABLE}
-DPERL_EXECUTABLE=${PERL_EXECUTABLE} -P
"${AOM_ROOT}/build/cmake/version.cmake"
COMMENT "Updating version info if necessary." VERBATIM)
COMMENT "Updating version info if necessary."
VERBATIM)
add_dependencies(aom_version aom_version_check)
# TODO(tomfinegan): Move rtcd target setup where it belongs for each rtcd
@ -236,7 +241,7 @@ endforeach()
# other pieces of the util support without defining usage_exit().
file(WRITE "${AOM_GEN_SRC_DIR}/usage_exit.c" "void usage_exit(void) {}")
file(WRITE "${AOM_GEN_SRC_DIR}/usage_exit.cc"
"extern \"C\" void usage_exit(void) {}")
"extern \"C\" void usage_exit(void) {}")
#
# Application and application support targets.
@ -255,31 +260,31 @@ endif()
if((CONFIG_AV1_DECODER OR CONFIG_AV1_ENCODER) AND ENABLE_EXAMPLES)
add_executable(resize_util "${AOM_ROOT}/examples/resize_util.c"
$<TARGET_OBJECTS:aom_common_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>)
list(APPEND AOM_APP_TARGETS resize_util)
endif()
if(CONFIG_AV1_DECODER AND ENABLE_EXAMPLES)
add_executable(aomdec "${AOM_ROOT}/apps/aomdec.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
add_executable(decode_to_md5 "${AOM_ROOT}/examples/decode_to_md5.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
add_executable(decode_with_drops "${AOM_ROOT}/examples/decode_with_drops.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
add_executable(simple_decoder "${AOM_ROOT}/examples/simple_decoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
add_executable(scalable_decoder "${AOM_ROOT}/examples/scalable_decoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
if(CONFIG_ANALYZER)
add_executable(analyzer "${AOM_ROOT}/examples/analyzer.cc"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
target_link_libraries(analyzer ${AOM_LIB_LINK_TYPE} ${wxWidgets_LIBRARIES})
list(APPEND AOM_APP_TARGETS analyzer)
list(APPEND AOM_DECODER_EXAMPLE_TARGETS analyzer)
@ -287,8 +292,8 @@ if(CONFIG_AV1_DECODER AND ENABLE_EXAMPLES)
if(CONFIG_INSPECTION)
add_executable(inspect "${AOM_ROOT}/examples/inspect.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
list(APPEND AOM_DECODER_EXAMPLE_TARGETS inspect)
if(EMSCRIPTEN)
@ -315,8 +320,8 @@ if(CONFIG_AV1_DECODER AND ENABLE_EXAMPLES)
endif()
# Maintain a list of decoder example targets.
list(APPEND AOM_DECODER_EXAMPLE_TARGETS aomdec decode_to_md5
decode_with_drops scalable_decoder simple_decoder)
list(APPEND AOM_DECODER_EXAMPLE_TARGETS aomdec decode_to_md5 decode_with_drops
scalable_decoder simple_decoder)
# Add decoder examples to the app targets list.
list(APPEND AOM_APP_TARGETS ${AOM_DECODER_EXAMPLE_TARGETS})
@ -325,31 +330,36 @@ endif()
if(CONFIG_AV1_ENCODER)
if(ENABLE_EXAMPLES)
add_executable(aomenc "${AOM_ROOT}/apps/aomenc.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>
$<TARGET_OBJECTS:aom_encoder_stats>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>
$<TARGET_OBJECTS:aom_encoder_stats>)
add_executable(lossless_encoder "${AOM_ROOT}/examples/lossless_encoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
add_executable(set_maps "${AOM_ROOT}/examples/set_maps.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
add_executable(simple_encoder "${AOM_ROOT}/examples/simple_encoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
add_executable(twopass_encoder "${AOM_ROOT}/examples/twopass_encoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
add_executable(noise_model "${AOM_ROOT}/examples/noise_model.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
add_executable(scalable_encoder "${AOM_ROOT}/examples/scalable_encoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
add_executable(svc_encoder_rtc "${AOM_ROOT}/examples/svc_encoder_rtc.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
# Maintain a list of encoder example targets.
list(APPEND AOM_ENCODER_EXAMPLE_TARGETS aomenc lossless_encoder noise_model
set_maps simple_encoder scalable_encoder twopass_encoder)
set_maps simple_encoder scalable_encoder twopass_encoder
svc_encoder_rtc)
endif()
if(ENABLE_TOOLS)
@ -358,7 +368,8 @@ if(CONFIG_AV1_ENCODER)
# TODO(tomfinegan): Sort out why a simple link command with
# aom_entropy_optimizer.c won't work on macos, but dragging in all the
# helper machinery allows the link to succeed.
add_executable(aom_entropy_optimizer "${AOM_GEN_SRC_DIR}/usage_exit.c"
add_executable(aom_entropy_optimizer
"${AOM_GEN_SRC_DIR}/usage_exit.c"
"${AOM_ROOT}/tools/aom_entropy_optimizer.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
@ -385,11 +396,11 @@ endif()
if(ENABLE_TOOLS)
if(CONFIG_AV1_DECODER)
add_executable(dump_obu "${AOM_GEN_SRC_DIR}/usage_exit.cc"
"${AOM_ROOT}/tools/dump_obu.cc"
"${AOM_ROOT}/tools/obu_parser.cc"
"${AOM_ROOT}/tools/obu_parser.h"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
"${AOM_ROOT}/tools/dump_obu.cc"
"${AOM_ROOT}/tools/obu_parser.cc"
"${AOM_ROOT}/tools/obu_parser.h"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
list(APPEND AOM_TOOL_TARGETS dump_obu)
list(APPEND AOM_APP_TARGETS dump_obu)
@ -404,16 +415,16 @@ endif()
if(ENABLE_EXAMPLES AND CONFIG_AV1_DECODER AND CONFIG_AV1_ENCODER)
add_executable(aom_cx_set_ref "${AOM_ROOT}/examples/aom_cx_set_ref.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
list(APPEND AOM_EXAMPLE_TARGETS aom_cx_set_ref)
list(APPEND AOM_APP_TARGETS aom_cx_set_ref)
endif()
if(ENABLE_EXAMPLES AND CONFIG_AV1_ENCODER)
add_executable(lightfield_encoder "${AOM_ROOT}/examples/lightfield_encoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_encoder_app_util>)
list(APPEND AOM_EXAMPLE_TARGETS lightfield_encoder)
list(APPEND AOM_APP_TARGETS lightfield_encoder)
endif()
@ -429,8 +440,8 @@ endif()
if(ENABLE_EXAMPLES AND CONFIG_AV1_DECODER)
add_executable(lightfield_decoder "${AOM_ROOT}/examples/lightfield_decoder.c"
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
$<TARGET_OBJECTS:aom_common_app_util>
$<TARGET_OBJECTS:aom_decoder_app_util>)
list(APPEND AOM_EXAMPLE_TARGETS lightfield_decoder)
list(APPEND AOM_APP_TARGETS lightfield_decoder)
endif()
@ -591,7 +602,8 @@ if(NOT AOM_DIST_DIR)
endif()
add_custom_target(dist
COMMAND ${CMAKE_COMMAND} -DAOM_ROOT=${AOM_ROOT}
COMMAND ${CMAKE_COMMAND}
-DAOM_ROOT=${AOM_ROOT}
-DAOM_CONFIG_DIR=${AOM_CONFIG_DIR}
-DAOM_DIST_DIR=${AOM_DIST_DIR}
-DAOM_DIST_APPS="${AOM_DIST_APPS}"
@ -612,8 +624,8 @@ endif()
# Collect all variables containing libaom source files.
get_cmake_property(all_cmake_vars VARIABLES)
foreach(var ${all_cmake_vars})
if("${var}" MATCHES "SOURCES$\|_INTRIN_\|_ASM_" AND NOT "${var}" MATCHES
"_APP_\|DOXYGEN\|LIBWEBM\|LIBYUV\|_PKG_\|TEST")
if("${var}" MATCHES "SOURCES$\|_INTRIN_\|_ASM_"
AND NOT "${var}" MATCHES "_APP_\|DOXYGEN\|LIBWEBM\|LIBYUV\|_PKG_\|TEST")
list(APPEND aom_source_vars ${var})
endif()
endforeach()
@ -632,9 +644,8 @@ foreach(aom_source_var ${aom_source_vars})
endforeach()
endforeach()
file(APPEND
"${libaom_srcs_txt_file}"
"# Files below this line are generated by the libaom build system.\n")
file(APPEND "${libaom_srcs_txt_file}"
"# Files below this line are generated by the libaom build system.\n")
foreach(aom_source_var ${aom_source_vars})
foreach(file ${${aom_source_var}})
if("${file}" MATCHES "${AOM_CONFIG_DIR}")
@ -667,15 +678,14 @@ foreach(aom_source_var ${aom_source_vars})
endif()
endforeach()
file(APPEND
"${libaom_srcs_gni_file}"
"\n# Files below this line are generated by the libaom build system.\n")
file(APPEND "${libaom_srcs_gni_file}"
"\n# Files below this line are generated by the libaom build system.\n")
foreach(aom_source_var ${aom_source_vars})
if("${${aom_source_var}}" MATCHES "${AOM_CONFIG_DIR}")
string(TOLOWER ${aom_source_var} aom_source_var_lowercase)
file(APPEND "${libaom_srcs_gni_file}"
"\n${aom_source_var_lowercase}_gen = [\n")
"\n${aom_source_var_lowercase}_gen = [\n")
endif()
foreach(file ${${aom_source_var}})
if(NOT "${file}" MATCHES "${AOM_ROOT}")

13
third_party/aom/aom/aom_codec.h vendored
View file

@ -503,19 +503,6 @@ typedef enum {
*/
const char *aom_obu_type_to_string(OBU_TYPE type);
/*!\brief Config Options
*
* This type allows to enumerate and control options defined for control
* via config file at runtime.
*/
typedef struct cfg_options {
/*!\brief Reflects if ext_partition should be enabled
*
* If this value is non-zero it enabled the feature
*/
unsigned int ext_partition;
} cfg_options_t;
/*!@} - end defgroup codec*/
#ifdef __cplusplus
}

3
third_party/aom/aom/aom_decoder.h vendored
View file

@ -42,7 +42,7 @@ extern "C" {
* fields to structures
*/
#define AOM_DECODER_ABI_VERSION \
(3 + AOM_CODEC_ABI_VERSION) /**<\hideinitializer*/
(4 + AOM_CODEC_ABI_VERSION) /**<\hideinitializer*/
/*! \brief Decoder capabilities bitfield
*
@ -98,7 +98,6 @@ typedef struct aom_codec_dec_cfg {
unsigned int w; /**< Width */
unsigned int h; /**< Height */
unsigned int allow_lowbitdepth; /**< Allow use of low-bitdepth coding path */
cfg_options_t cfg; /**< Options defined per config attributes */
} aom_codec_dec_cfg_t; /**< alias for struct aom_codec_dec_cfg */
/*!\brief Initialize a decoder instance

9
third_party/aom/aom/aom_encoder.h vendored
View file

@ -41,7 +41,7 @@ extern "C" {
* fields to structures
*/
#define AOM_ENCODER_ABI_VERSION \
(5 + AOM_CODEC_ABI_VERSION) /**<\hideinitializer*/
(6 + AOM_CODEC_ABI_VERSION) /**<\hideinitializer*/
/*! \brief Encoder capabilities bitfield
*
@ -74,7 +74,7 @@ extern "C" {
* This structure is able to hold a reference to any fixed size buffer.
*/
typedef struct aom_fixed_buf {
void *buf; /**< Pointer to the data */
void *buf; /**< Pointer to the data. Does NOT own the data! */
size_t sz; /**< Length of the buffer, in chars */
} aom_fixed_buf_t; /**< alias for struct aom_fixed_buf */
@ -723,11 +723,6 @@ typedef struct aom_codec_enc_cfg {
* The number of heights specified is given by tile_height_count
*/
int tile_heights[MAX_TILE_HEIGHTS];
/*!\brief Options defined per config file
*
*/
cfg_options_t cfg;
} aom_codec_enc_cfg_t; /**< alias for struct aom_codec_enc_cfg */
/*!\brief Initialize an encoder instance

76
third_party/aom/aom/aom_image.h vendored
View file

@ -30,7 +30,7 @@ extern "C" {
* types, removing or reassigning enums, adding/removing/rearranging
* fields to structures
*/
#define AOM_IMAGE_ABI_VERSION (5) /**<\hideinitializer*/
#define AOM_IMAGE_ABI_VERSION (6) /**<\hideinitializer*/
#define AOM_IMG_FMT_PLANAR 0x100 /**< Image is a planar format. */
#define AOM_IMG_FMT_UV_FLIP 0x200 /**< V plane precedes U in memory. */
@ -137,6 +137,16 @@ typedef enum aom_chroma_sample_position {
AOM_CSP_RESERVED = 3 /**< Reserved value */
} aom_chroma_sample_position_t; /**< alias for enum aom_transfer_function */
/*!\brief Array of aom_metadata structs for an image. */
typedef struct aom_metadata_array aom_metadata_array_t;
/*!\brief Metadata payload. */
typedef struct aom_metadata {
uint8_t type; /**< Metadata type */
uint8_t *payload; /**< Metadata payload data */
size_t sz; /**< Metadata payload size */
} aom_metadata_t;
/**\brief Image Descriptor */
typedef struct aom_image {
aom_img_fmt_t fmt; /**< Image Format */
@ -188,6 +198,9 @@ typedef struct aom_image {
int img_data_owner; /**< private */
int self_allocd; /**< private */
aom_metadata_array_t
*metadata; /**< Metadata payloads associated with the image. */
void *fb_priv; /**< Frame buffer data associated with the image. */
} aom_image_t; /**< alias for struct aom_image */
@ -202,7 +215,7 @@ typedef struct aom_image_rect {
/*!\brief Open a descriptor, allocating storage for the underlying image
*
* Returns a descriptor for storing an image of the given format. The
* storage for the descriptor is allocated on the heap.
* storage for the image is allocated on the heap.
*
* \param[in] img Pointer to storage for descriptor. If this parameter
* is NULL, the storage for the descriptor will be
@ -211,7 +224,7 @@ typedef struct aom_image_rect {
* \param[in] d_w Width of the image
* \param[in] d_h Height of the image
* \param[in] align Alignment, in bytes, of the image buffer and
* each row in the image(stride).
* each row in the image (stride).
*
* \return Returns a pointer to the initialized image descriptor. If the img
* parameter is non-null, the value of the img parameter will be
@ -224,7 +237,7 @@ aom_image_t *aom_img_alloc(aom_image_t *img, aom_img_fmt_t fmt,
/*!\brief Open a descriptor, using existing storage for the underlying image
*
* Returns a descriptor for storing an image of the given format. The
* storage for descriptor has been allocated elsewhere, and a descriptor is
* storage for the image has been allocated elsewhere, and a descriptor is
* desired to "wrap" that storage.
*
* \param[in] img Pointer to storage for descriptor. If this parameter
@ -233,7 +246,8 @@ aom_image_t *aom_img_alloc(aom_image_t *img, aom_img_fmt_t fmt,
* \param[in] fmt Format for the image
* \param[in] d_w Width of the image
* \param[in] d_h Height of the image
* \param[in] align Alignment, in bytes, of each row in the image.
* \param[in] align Alignment, in bytes, of each row in the image
* (stride).
* \param[in] img_data Storage to use for the image
*
* \return Returns a pointer to the initialized image descriptor. If the img
@ -248,7 +262,7 @@ aom_image_t *aom_img_wrap(aom_image_t *img, aom_img_fmt_t fmt, unsigned int d_w,
* border
*
* Returns a descriptor for storing an image of the given format and its
* borders. The storage for the descriptor is allocated on the heap.
* borders. The storage for the image is allocated on the heap.
*
* \param[in] img Pointer to storage for descriptor. If this parameter
* is NULL, the storage for the descriptor will be
@ -257,8 +271,8 @@ aom_image_t *aom_img_wrap(aom_image_t *img, aom_img_fmt_t fmt, unsigned int d_w,
* \param[in] d_w Width of the image
* \param[in] d_h Height of the image
* \param[in] align Alignment, in bytes, of the image buffer and
* each row in the image(stride).
* \param[in] size_align Alignment, in bytes, of the image width and height.
* each row in the image (stride).
* \param[in] size_align Alignment, in pixels, of the image width and height.
* \param[in] border A border that is padded on four sides of the image.
*
* \return Returns a pointer to the initialized image descriptor. If the img
@ -323,6 +337,52 @@ int aom_img_plane_width(const aom_image_t *img, int plane);
*/
int aom_img_plane_height(const aom_image_t *img, int plane);
/*!\brief Add metadata to image.
*
* Adds metadata to aom_image_t.
* Function makes a copy of the provided data parameter.
*
* \param[in] img Image descriptor
* \param[in] type Metadata type
* \param[in] data Metadata contents
* \param[in] sz Metadata contents size
*/
int aom_img_add_metadata(aom_image_t *img, uint8_t type, uint8_t *data,
size_t sz);
/*!\brief Remove metadata from image.
*
* Removes all metadata in image metadata list and sets metadata list pointer
* to NULL.
* Returns the number of deleted metadata structs.
*
* \param[in] img Image descriptor
*/
size_t aom_img_remove_metadata(aom_image_t *img);
/*!\brief Allocate memory for aom_metadata struct.
*
* Allocates memory for aom_metadata struct and sets its type. Optionally
* allocates storage for the metadata payload and copies the payload data
* into the aom_metadata struct:
* - When sz is > 0 and data is NULL, allocates metadata payload buffer of sz.
* - When sz is > 0 and data is non-NULL, a metadata payload buffer of sz
* is allocated and sz bytes are copied from data into the payload buffer.
*
* \param[in] type Metadata type
* \param[in] data Metadata data pointer
* \param[in] sz Metadata size
*/
aom_metadata_t *aom_img_metadata_alloc(uint8_t type, uint8_t *data, size_t sz);
/*!\brief Free metadata struct.
*
* Free metadata struct and its buffer.
*
* \param[in] metadata Metadata struct pointer
*/
int aom_img_metadata_free(aom_metadata_t *metadata);
#ifdef __cplusplus
} // extern "C"
#endif

143
third_party/aom/aom/aomcx.h vendored
View file

@ -173,9 +173,9 @@ enum aome_enc_control_id {
/*!\brief Codec control function to set encoder internal speed settings.
*
* Changes in this value influences, among others, the encoder's selection
* of motion estimation methods. Values greater than 0 will increase encoder
* speed at the expense of quality.
* Changes in this value influences the complexity of algorithms used in
* encoding process, values greater than 0 will increase encoder speed at
* the expense of quality.
*
* \note Valid range: 0..8
*/
@ -218,10 +218,10 @@ enum aome_enc_control_id {
*/
AOME_SET_TUNING = AOME_SET_ARNR_STRENGTH + 2,
/*!\brief Codec control function to set constrained quality level.
/*!\brief Codec control function to set constrained / constant quality level.
*
* \attention For this value to be used aom_codec_enc_cfg_t::rc_end_usage
* must be set to #AOM_CQ.
* must be set to #AOM_CQ or #AOM_Q.
* \note Valid range: 0..63
*/
AOME_SET_CQ_LEVEL,
@ -328,6 +328,12 @@ enum aome_enc_control_id {
*/
AV1E_SET_ENABLE_TPL_MODEL,
/*!\brief Codec control function to enable temporal filtering on key frame.
*
* By default, this feature is on.
*/
AV1E_SET_ENABLE_KEYFRAME_FILTERING,
/*!\brief Codec control function to enable frame parallel decoding feature.
*
* AV1 has a bitstream feature to reduce decoding dependency between frames
@ -518,8 +524,8 @@ enum aome_enc_control_id {
* 4 digits).
* AB: OP index.
* xy: Target level index for the OP. Can be values 0~23(corresponding to
* level 2.0 ~ 7.3) or 31(maximum level parameter, no level-based
* constraints).
* level 2.0 ~ 7.3) or 24(keep level stats only for level monitoring) or
* 31(maximum level parameter, no level-based constraints).
* E.g. "0" means target level index 0 for the 0th OP;
* "111" means target level index 11 for the 1st OP;
* "1021" means target level index 21 for the 10th OP.
@ -528,7 +534,9 @@ enum aome_enc_control_id {
*/
AV1E_SET_TARGET_SEQ_LEVEL_IDX,
/*!\brief Codec control function to get sequence level index.
/*!\brief Codec control function to get sequence level index for each
* operating point. There can be at most 32 operating points. The results will
* be written into a provided integer array of sufficient size.
*/
AV1E_GET_SEQ_LEVEL_IDX,
@ -570,6 +578,16 @@ enum aome_enc_control_id {
*/
AV1E_SET_ENABLE_RESTORATION,
/*!\brief Codec control function to force video mode
*
* 0 = do not force video mode
* 1 = force video mode even for a single frame
*
* By default, the encoder does not force video and allows still picture.
*
*/
AV1E_SET_FORCE_VIDEO_MODE,
/*!\brief Codec control function to predict with OBMC mode.
*
* 0 = do not allow OBMC mode
@ -851,6 +869,15 @@ enum aome_enc_control_id {
*/
AV1E_SET_ENABLE_DUAL_FILTER,
/*!\brief Codec control function to turn on / off delta quantization in chroma
* planes usage for a sequence.
*
* This will enable or disable use of chroma deltaq.
* The default value is 0.
*
*/
AV1E_SET_ENABLE_CHROMA_DELTAQ,
/*!\brief Codec control function to turn on / off masked compound usage
* for a sequence.
*
@ -979,6 +1006,16 @@ enum aome_enc_control_id {
*/
AV1E_SET_ENABLE_SUPERRES,
/*!\brief Codec control function to turn on / off overlay frames for
* fitlered ALTREF frames.
*
* This will enable or disable coding of overlay frames for filtered ALTREF
* frames. When set to 0, overlay frames are not used but show existing frame
* is used to display the filtered ALTREF frame as is. As a result the decoded
* frame rate remains the same as the display frame rate. The default is 1.
*/
AV1E_SET_ENABLE_OVERLAY,
/*!\brief Codec control function to turn on/off palette mode */
AV1E_SET_ENABLE_PALETTE,
@ -1095,11 +1132,39 @@ enum aome_enc_control_id {
*/
AV1E_SET_MODE_COST_UPD_FREQ,
/*!\brief Control to set frequency of the cost updates for motion vectors
* Possible values are:
* 0: Update at SB level (default)
* 1: Update at SB row level in tile
* 2: Update at tile level
*/
AV1E_SET_MV_COST_UPD_FREQ,
/*!\brief Control to set bit mask that specifies which tier each of the 32
* possible operating points conforms to.
* Bit value 0: Main Tier; 1: High Tier.
*/
AV1E_SET_TIER_MASK,
/*!\brief Control to set minimum compression ratio.
* Take integer values. If non-zero, encoder will try to keep the compression
* ratio of each frame to be higher than the given value divided by 100.
* E.g. 850 means minimum compression ratio of 8.5.
*/
AV1E_SET_MIN_CR,
/*!\brief Codec control function to set the layer id.
*/
AV1E_SET_SVC_LAYER_ID = 150,
/*!\brief Codec control function to set SVC paramaeters.
*/
AV1E_SET_SVC_PARAMS = 151,
/*!\brief Codec control function to set reference frame config:
* the ref_idx and the refresh flags for each buffer slot.
*/
AV1E_SET_SVC_REF_FRAME_CONFIG = 152
};
/*!\brief aom 1-D scaling mode
@ -1189,6 +1254,39 @@ typedef enum {
AOM_TUNE_DAALA_DIST
} aom_tune_metric;
#define AOM_MAX_LAYERS 32 /**< Max number of layers */
#define AOM_MAX_SS_LAYERS 4 /**< Max number of spatial layers */
#define AOM_MAX_TS_LAYERS 8 /**< Max number of temporal layers */
/*!brief Struct for spatial and temporal layer ID */
typedef struct aom_svc_layer_id {
int spatial_layer_id; /**< Spatial layer ID */
int temporal_layer_id; /**< Temporal layer ID */
} aom_svc_layer_id_t;
/*!brief Parameter type for SVC */
typedef struct aom_svc_params {
int number_spatial_layers; /**< Number of spatial layers */
int number_temporal_layers; /**< Number of temporal layers */
int max_quantizers[AOM_MAX_LAYERS]; /**< Max Q for each layer */
int min_quantizers[AOM_MAX_LAYERS]; /**< Min Q for each layer */
int scaling_factor_num[AOM_MAX_SS_LAYERS]; /**< Scaling factor-numerator */
int scaling_factor_den[AOM_MAX_SS_LAYERS]; /**< Scaling factor-denominator */
/*! Target bitrate for each layer */
int layer_target_bitrate[AOM_MAX_LAYERS];
/*! Frame rate factor for each temporal layer */
int framerate_factor[AOM_MAX_TS_LAYERS];
} aom_svc_params_t;
/*!brief Parameters for setting ref frame config */
typedef struct aom_svc_ref_frame_config {
// 7 references: LAST_FRAME (0), LAST2_FRAME(1), LAST3_FRAME(2),
// GOLDEN_FRAME(3), BWDREF_FRAME(4), ALTREF2_FRAME(5), ALTREF_FRAME(6).
/*! Buffer slot index for each of 7 references. */
int ref_idx[7];
int refresh[8]; /**< Refresh flag for each of the 8 slots. */
} aom_svc_ref_frame_config_t;
/*!\cond */
/*!\brief Encoder control function parameter type
*
@ -1242,6 +1340,9 @@ AOM_CTRL_USE_TYPE(AV1E_SET_TILE_ROWS, unsigned int)
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_TPL_MODEL, unsigned int)
#define AOM_CTRL_AV1E_SET_ENABLE_TPL_MODEL
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_KEYFRAME_FILTERING, unsigned int)
#define AOM_CTRL_AV1E_SET_ENABLE_KEYFRAME_FILTERING
AOM_CTRL_USE_TYPE(AOME_GET_LAST_QUANTIZER, int *)
#define AOM_CTRL_AOME_GET_LAST_QUANTIZER
AOM_CTRL_USE_TYPE(AOME_GET_LAST_QUANTIZER_64, int *)
@ -1267,6 +1368,9 @@ AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_CDEF, unsigned int)
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_RESTORATION, unsigned int)
#define AOM_CTRL_AV1E_SET_ENABLE_RESTORATION
AOM_CTRL_USE_TYPE(AV1E_SET_FORCE_VIDEO_MODE, unsigned int)
#define AOM_CTRL_AV1E_SET_FORCE_VIDEO_MODE
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_OBMC, unsigned int)
#define AOM_CTRL_AV1E_SET_ENABLE_OBMC
@ -1344,6 +1448,9 @@ AOM_CTRL_USE_TYPE(AV1E_SET_ALLOW_REF_FRAME_MVS, int)
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_DUAL_FILTER, int)
#define AOM_CTRL_AV1E_SET_ENABLE_DUAL_FILTER
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_CHROMA_DELTAQ, int)
#define AOM_CTRL_AV1E_SET_ENABLE_CHROMA_DELTAQ
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_MASKED_COMP, int)
#define AOM_CTRL_AV1E_SET_ENABLE_MASKED_COMP
@ -1389,6 +1496,9 @@ AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_CFL_INTRA, int)
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_SUPERRES, int)
#define AOM_CTRL_AV1E_SET_ENABLE_SUPERRES
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_OVERLAY, int)
#define AOM_CTRL_AV1E_SET_ENABLE_OVERLAY
AOM_CTRL_USE_TYPE(AV1E_SET_ENABLE_PALETTE, int)
#define AOM_CTRL_AV1E_SET_ENABLE_PALETTE
@ -1473,13 +1583,11 @@ AOM_CTRL_USE_TYPE(AV1E_SET_FILM_GRAIN_TABLE, const char *)
AOM_CTRL_USE_TYPE(AV1E_SET_CDF_UPDATE_MODE, unsigned int)
#define AOM_CTRL_AV1E_SET_CDF_UPDATE_MODE
#ifdef CONFIG_DENOISE
AOM_CTRL_USE_TYPE(AV1E_SET_DENOISE_NOISE_LEVEL, int);
#define AOM_CTRL_AV1E_SET_DENOISE_NOISE_LEVEL
AOM_CTRL_USE_TYPE(AV1E_SET_DENOISE_BLOCK_SIZE, unsigned int);
#define AOM_CTRL_AV1E_SET_DENOISE_BLOCK_SIZE
#endif
AOM_CTRL_USE_TYPE(AV1E_SET_CHROMA_SUBSAMPLING_X, unsigned int)
#define AOM_CTRL_AV1E_SET_CHROMA_SUBSAMPLING_X
@ -1517,12 +1625,27 @@ AOM_CTRL_USE_TYPE(AV1E_SET_COEFF_COST_UPD_FREQ, unsigned int)
AOM_CTRL_USE_TYPE(AV1E_SET_MODE_COST_UPD_FREQ, unsigned int)
#define AOM_CTRL_AV1E_SET_MODE_COST_UPD_FREQ
AOM_CTRL_USE_TYPE(AV1E_SET_MV_COST_UPD_FREQ, unsigned int)
#define AOM_CTRL_AV1E_SET_MV_COST_UPD_FREQ
AOM_CTRL_USE_TYPE(AV1E_SET_TARGET_SEQ_LEVEL_IDX, int)
#define AOM_CTRL_AV1E_SET_TARGET_SEQ_LEVEL_IDX
AOM_CTRL_USE_TYPE(AV1E_SET_TIER_MASK, unsigned int)
#define AOM_CTRL_AV1E_SET_TIER_MASK
AOM_CTRL_USE_TYPE(AV1E_SET_MIN_CR, unsigned int)
#define AOM_CTRL_AV1E_SET_MIN_CR
AOM_CTRL_USE_TYPE(AV1E_SET_SVC_LAYER_ID, aom_svc_layer_id_t *)
#define AOME_CTRL_AV1E_SET_SVC_LAYER_ID
AOM_CTRL_USE_TYPE(AV1E_SET_SVC_PARAMS, aom_svc_params_t *)
#define AOME_CTRL_AV1E_SET_SVC_PARAMS
AOM_CTRL_USE_TYPE(AV1E_SET_SVC_REF_FRAME_CONFIG, aom_svc_ref_frame_config_t *)
#define AOME_CTRL_AV1E_SET_SVC_REF_FRAME_CONFIG
/*!\endcond */
/*! @} - end defgroup aom_encoder */
#ifdef __cplusplus

6
third_party/aom/aom/exports_com vendored
View file

@ -10,12 +10,18 @@ text aom_codec_version
text aom_codec_version_extra_str
text aom_codec_version_str
text aom_free
text aom_img_add_metadata
text aom_img_alloc
text aom_img_alloc_with_border
text aom_img_flip
text aom_img_free
text aom_img_metadata_array_free
text aom_img_metadata_array_alloc
text aom_img_metadata_free
text aom_img_metadata_alloc
text aom_img_plane_height
text aom_img_plane_width
text aom_img_remove_metadata
text aom_img_set_rect
text aom_img_wrap
text aom_malloc

2
third_party/aom/aom/exports_test vendored
View file

@ -1,2 +1,4 @@
text aom_copy_metadata_to_frame_buffer
text aom_dsp_rtcd
text aom_remove_metadata_from_frame_buffer
text aom_scale_rtcd

94
third_party/aom/aom/internal/aom_image_internal.h vendored Normal file
View file

@ -0,0 +1,94 @@
/*
* Copyright (c) 2019, Alliance for Open Media. All rights reserved
*
* This source code is subject to the terms of the BSD 2 Clause License and
* the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
* was not distributed with this source code in the LICENSE file, you can
* obtain it at www.aomedia.org/license/software. If the Alliance for Open
* Media Patent License 1.0 was not distributed with this source code in the
* PATENTS file, you can obtain it at www.aomedia.org/license/patent.
*/
/*!\file
* \brief Describes the internal functions associated with the aom image
* descriptor.
*
*/
#ifndef AOM_AOM_INTERNAL_AOM_IMAGE_INTERNAL_H_
#define AOM_AOM_INTERNAL_AOM_IMAGE_INTERNAL_H_
#include "aom/aom_image.h"
#ifdef __cplusplus
extern "C" {
#endif
/*!\brief Array of aom_metadata structs for an image. */
struct aom_metadata_array {
size_t sz; /* Number of metadata structs in the list */
aom_metadata_t **metadata_array; /* Array of metadata structs */
};
/*!\brief Alloc memory for aom_metadata_array struct.
*
* Allocate memory for aom_metadata_array struct.
* If sz is 0 the aom_metadata_array structs internal buffer list will be NULL,
* but the aom_metadata_array struct itself will still be allocated.
* Returns a pointer to the allocated struct or NULL on failure.
*
* \param[in] sz Size of internal metadata list buffer
*/
aom_metadata_array_t *aom_img_metadata_array_alloc(size_t sz);
/*!\brief Free metadata array struct.
*
* Free metadata array struct and all metadata structs inside.
* Returns the number of deleted metadata structs.
*
* \param[in] arr Metadata array struct pointer
*/
size_t aom_img_metadata_array_free(aom_metadata_array_t *arr);
typedef void *(*aom_alloc_img_data_cb_fn_t)(void *priv, size_t size);
/*!\brief Open a descriptor, allocating storage for the underlying image by
* using the provided callback function.
*
* Returns a descriptor for storing an image of the given format. The storage
* for the image is allocated by using the provided callback function. Unlike
* aom_img_alloc(), the returned descriptor does not own the storage for the
* image. The caller is responsible for freeing the storage for the image.
*
* Note: If the callback function is invoked and succeeds,
* aom_img_alloc_with_cb() is guaranteed to succeed. Therefore, if
* aom_img_alloc_with_cb() fails, the caller is assured that no storage was
* allocated.
*
* \param[in] img Pointer to storage for descriptor. If this parameter
* is NULL, the storage for the descriptor will be
* allocated on the heap.
* \param[in] fmt Format for the image
* \param[in] d_w Width of the image
* \param[in] d_h Height of the image
* \param[in] align Alignment, in bytes, of the image buffer and
* each row in the image (stride).
* \param[in] alloc_cb Callback function used to allocate storage for the
* image.
* \param[in] cb_priv The first argument ('priv') for the callback
* function.
*
* \return Returns a pointer to the initialized image descriptor. If the img
* parameter is non-null, the value of the img parameter will be
* returned.
*/
aom_image_t *aom_img_alloc_with_cb(aom_image_t *img, aom_img_fmt_t fmt,
unsigned int d_w, unsigned int d_h,
unsigned int align,
aom_alloc_img_data_cb_fn_t alloc_cb,
void *cb_priv);
#ifdef __cplusplus
} // extern "C"
#endif
#endif // AOM_AOM_INTERNAL_AOM_IMAGE_INTERNAL_H_

20
third_party/aom/aom/src/aom_encoder.c vendored
View file

@ -144,12 +144,12 @@ aom_codec_err_t aom_codec_enc_init_multi_ver(
aom_codec_err_t aom_codec_enc_config_default(aom_codec_iface_t *iface,
aom_codec_enc_cfg_t *cfg,
unsigned int usage) {
unsigned int reserved) {
aom_codec_err_t res;
aom_codec_enc_cfg_map_t *map;
int i;
if (!iface || !cfg || usage > INT_MAX)
if (!iface || !cfg || reserved > INT_MAX)
res = AOM_CODEC_INVALID_PARAM;
else if (!(iface->caps & AOM_CODEC_CAP_ENCODER))
res = AOM_CODEC_INCAPABLE;
@ -158,20 +158,15 @@ aom_codec_err_t aom_codec_enc_config_default(aom_codec_iface_t *iface,
for (i = 0; i < iface->enc.cfg_map_count; ++i) {
map = iface->enc.cfg_maps + i;
if (map->usage == (int)usage) {
if (map->usage == (int)reserved) {
*cfg = map->cfg;
cfg->g_usage = usage;
cfg->g_usage = reserved;
res = AOM_CODEC_OK;
break;
}
}
}
/* default values */
if (cfg) {
cfg->cfg.ext_partition = 1;
}
return res;
}
@ -190,8 +185,11 @@ aom_codec_err_t aom_codec_enc_config_default(aom_codec_iface_t *iface,
#if HAVE_FEXCEPT && CONFIG_DEBUG
#define FLOATING_POINT_SET_EXCEPTIONS \
const int float_excepts = feenableexcept(FE_DIVBYZERO);
#define FLOATING_POINT_RESTORE_EXCEPTIONS feenableexcept(float_excepts);
const int float_excepts = \
feenableexcept(FE_DIVBYZERO | FE_UNDERFLOW | FE_OVERFLOW);
#define FLOATING_POINT_RESTORE_EXCEPTIONS \
fedisableexcept(FE_ALL_EXCEPT); \
feenableexcept(float_excepts);
#else
#define FLOATING_POINT_SET_EXCEPTIONS
#define FLOATING_POINT_RESTORE_EXCEPTIONS

162
third_party/aom/aom/src/aom_image.c vendored
View file

@ -14,6 +14,7 @@
#include "aom/aom_image.h"
#include "aom/aom_integer.h"
#include "aom/internal/aom_image_internal.h"
#include "aom_mem/aom_mem.h"
static INLINE unsigned int align_image_dimension(unsigned int d,
@ -29,8 +30,12 @@ static INLINE unsigned int align_image_dimension(unsigned int d,
static aom_image_t *img_alloc_helper(
aom_image_t *img, aom_img_fmt_t fmt, unsigned int d_w, unsigned int d_h,
unsigned int buf_align, unsigned int stride_align, unsigned int size_align,
unsigned char *img_data, unsigned int border) {
unsigned int h, w, s, xcs, ycs, bps;
unsigned int border, unsigned char *img_data,
aom_alloc_img_data_cb_fn_t alloc_cb, void *cb_priv) {
/* NOTE: In this function, bit_depth is either 8 or 16 (if
* AOM_IMG_FMT_HIGHBITDEPTH is set), never 10 or 12.
*/
unsigned int h, w, s, xcs, ycs, bps, bit_depth;
unsigned int stride_in_bytes;
/* Treat align==0 like align==1 */
@ -57,15 +62,17 @@ static aom_image_t *img_alloc_helper(
case AOM_IMG_FMT_YV12:
case AOM_IMG_FMT_AOMI420:
case AOM_IMG_FMT_AOMYV12: bps = 12; break;
case AOM_IMG_FMT_I422:
case AOM_IMG_FMT_I422: bps = 16; break;
case AOM_IMG_FMT_I444: bps = 24; break;
case AOM_IMG_FMT_YV1216:
case AOM_IMG_FMT_I42016: bps = 24; break;
case AOM_IMG_FMT_I42216:
case AOM_IMG_FMT_I42216: bps = 32; break;
case AOM_IMG_FMT_I44416: bps = 48; break;
default: bps = 16; break;
}
bit_depth = (fmt & AOM_IMG_FMT_HIGHBITDEPTH) ? 16 : 8;
/* Get chroma shift values for this format */
switch (fmt) {
case AOM_IMG_FMT_I420:
@ -93,9 +100,9 @@ static aom_image_t *img_alloc_helper(
w = align_image_dimension(d_w, xcs, size_align);
h = align_image_dimension(d_h, ycs, size_align);
s = (fmt & AOM_IMG_FMT_PLANAR) ? w : bps * w / 8;
s = (fmt & AOM_IMG_FMT_PLANAR) ? w : bps * w / bit_depth;
s = (s + 2 * border + stride_align - 1) & ~(stride_align - 1);
stride_in_bytes = (fmt & AOM_IMG_FMT_HIGHBITDEPTH) ? s * 2 : s;
stride_in_bytes = s * bit_depth / 8;
/* Allocate the new image */
if (!img) {
@ -113,20 +120,29 @@ static aom_image_t *img_alloc_helper(
if (!img_data) {
const uint64_t alloc_size =
(fmt & AOM_IMG_FMT_PLANAR)
? (uint64_t)(h + 2 * border) * stride_in_bytes * bps / 8
? (uint64_t)(h + 2 * border) * stride_in_bytes * bps / bit_depth
: (uint64_t)(h + 2 * border) * stride_in_bytes;
if (alloc_size != (size_t)alloc_size) goto fail;
img->img_data = (uint8_t *)aom_memalign(buf_align, (size_t)alloc_size);
img->img_data_owner = 1;
if (alloc_cb) {
const size_t padded_alloc_size = (size_t)alloc_size + buf_align - 1;
img->img_data = (uint8_t *)alloc_cb(cb_priv, padded_alloc_size);
if (img->img_data) {
img->img_data = (uint8_t *)aom_align_addr(img->img_data, buf_align);
}
img->img_data_owner = 0;
} else {
img->img_data = (uint8_t *)aom_memalign(buf_align, (size_t)alloc_size);
img->img_data_owner = 1;
}
img->sz = (size_t)alloc_size;
}
if (!img->img_data) goto fail;
img->fmt = fmt;
img->bit_depth = (fmt & AOM_IMG_FMT_HIGHBITDEPTH) ? 16 : 8;
img->bit_depth = bit_depth;
// aligned width and aligned height
img->w = w;
img->h = h;
@ -138,8 +154,10 @@ static aom_image_t *img_alloc_helper(
img->stride[AOM_PLANE_Y] = stride_in_bytes;
img->stride[AOM_PLANE_U] = img->stride[AOM_PLANE_V] = stride_in_bytes >> xcs;
/* Default viewport to entire image */
if (!aom_img_set_rect(img, 0, 0, d_w, d_h, border)) return img;
/* Default viewport to entire image. (This aom_img_set_rect call always
* succeeds.) */
aom_img_set_rect(img, 0, 0, d_w, d_h, border);
return img;
fail:
aom_img_free(img);
@ -149,15 +167,26 @@ fail:
aom_image_t *aom_img_alloc(aom_image_t *img, aom_img_fmt_t fmt,
unsigned int d_w, unsigned int d_h,
unsigned int align) {
return img_alloc_helper(img, fmt, d_w, d_h, align, align, 1, NULL, 0);
return img_alloc_helper(img, fmt, d_w, d_h, align, align, 1, 0, NULL, NULL,
NULL);
}
aom_image_t *aom_img_alloc_with_cb(aom_image_t *img, aom_img_fmt_t fmt,
unsigned int d_w, unsigned int d_h,
unsigned int align,
aom_alloc_img_data_cb_fn_t alloc_cb,
void *cb_priv) {
return img_alloc_helper(img, fmt, d_w, d_h, align, align, 1, 0, NULL,
alloc_cb, cb_priv);
}
aom_image_t *aom_img_wrap(aom_image_t *img, aom_img_fmt_t fmt, unsigned int d_w,
unsigned int d_h, unsigned int stride_align,
unsigned char *img_data) {
/* By setting buf_align = 1, we don't change buffer alignment in this
* function. */
return img_alloc_helper(img, fmt, d_w, d_h, 1, stride_align, 1, img_data, 0);
/* Set buf_align = 1. It is ignored by img_alloc_helper because img_data is
* not NULL. */
return img_alloc_helper(img, fmt, d_w, d_h, 1, stride_align, 1, 0, img_data,
NULL, NULL);
}
aom_image_t *aom_img_alloc_with_border(aom_image_t *img, aom_img_fmt_t fmt,
@ -165,8 +194,8 @@ aom_image_t *aom_img_alloc_with_border(aom_image_t *img, aom_img_fmt_t fmt,
unsigned int align,
unsigned int size_align,
unsigned int border) {
return img_alloc_helper(img, fmt, d_w, d_h, align, align, size_align, NULL,
border);
return img_alloc_helper(img, fmt, d_w, d_h, align, align, size_align, border,
NULL, NULL, NULL);
}
int aom_img_set_rect(aom_image_t *img, unsigned int x, unsigned int y,
@ -238,6 +267,7 @@ void aom_img_flip(aom_image_t *img) {
void aom_img_free(aom_image_t *img) {
if (img) {
aom_img_remove_metadata(img);
if (img->img_data && img->img_data_owner) aom_free(img->img_data);
if (img->self_allocd) free(img);
@ -257,3 +287,99 @@ int aom_img_plane_height(const aom_image_t *img, int plane) {
else
return img->d_h;
}
aom_metadata_t *aom_img_metadata_alloc(uint8_t type, uint8_t *data, size_t sz) {
aom_metadata_t *metadata =
(aom_metadata_t *)calloc(1, sizeof(aom_metadata_t));
if (!metadata) return NULL;
metadata->type = type;
if (sz > 0) {
metadata->payload = (uint8_t *)calloc(sz, sizeof(uint8_t));
if (!metadata->payload) {
free(metadata);
return NULL;
}
if (data) {
memcpy(metadata->payload, data, sz);
metadata->sz = sz;
}
}
return metadata;
}
int aom_img_metadata_free(aom_metadata_t *metadata) {
if (!metadata) return -1;
if (metadata->payload) free(metadata->payload);
free(metadata);
return 0;
}
aom_metadata_array_t *aom_img_metadata_array_alloc(size_t sz) {
aom_metadata_array_t *arr =
(aom_metadata_array_t *)calloc(1, sizeof(aom_metadata_array_t));
if (!arr) return NULL;
if (sz > 0) {
arr->metadata_array =
(aom_metadata_t **)calloc(sz, sizeof(aom_metadata_t *));
if (!arr->metadata_array) {
aom_img_metadata_array_free(arr);
return NULL;
}
arr->sz = sz;
}
return arr;
}
size_t aom_img_metadata_array_free(aom_metadata_array_t *arr) {
size_t deleted_metadatas = 0;
if (!arr) return deleted_metadatas;
if (arr->metadata_array) {
for (size_t i = 0; i < arr->sz; i++) {
if (aom_img_metadata_free(arr->metadata_array[i]) == 0) {
deleted_metadatas++;
}
}
free(arr->metadata_array);
}
free(arr);
return deleted_metadatas;
}
int aom_img_add_metadata(aom_image_t *img, uint8_t type, uint8_t *data,
size_t sz) {
if (!img) return -1;
if (!img->metadata) {
img->metadata = aom_img_metadata_array_alloc(0);
if (!img->metadata) return -1;
}
aom_metadata_t *metadata = aom_img_metadata_alloc(type, data, sz);
if (!metadata) goto fail;
if (!img->metadata->metadata_array) {
img->metadata->metadata_array =
(aom_metadata_t **)calloc(1, sizeof(metadata));
if (!img->metadata->metadata_array || img->metadata->sz != 0) {
aom_img_metadata_free(metadata);
goto fail;
}
} else {
img->metadata->metadata_array =
(aom_metadata_t **)realloc(img->metadata->metadata_array,
(img->metadata->sz + 1) * sizeof(metadata));
}
img->metadata->metadata_array[img->metadata->sz] = metadata;
img->metadata->sz++;
return 0;
fail:
aom_img_metadata_array_free(img->metadata);
img->metadata = NULL;
return -1;
}
size_t aom_img_remove_metadata(aom_image_t *img) {
if (img && img->metadata) {
size_t sz = aom_img_metadata_array_free(img->metadata);
img->metadata = NULL;
return sz;
}
return 0;
}

2
third_party/aom/aom_dsp/aom_convolve.c vendored
View file

@ -130,6 +130,7 @@ void aom_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE int highbd_vert_scalar_product(const uint16_t *a,
ptrdiff_t a_stride,
const int16_t *b) {
@ -236,3 +237,4 @@ void aom_highbd_convolve_copy_c(const uint8_t *src8, ptrdiff_t src_stride,
dst += dst_stride;
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH

44
third_party/aom/aom_dsp/aom_dsp.cmake vendored
View file

@ -76,6 +76,12 @@ list(APPEND AOM_DSP_COMMON_INTRIN_SSE2
"${AOM_ROOT}/aom_dsp/x86/avg_intrin_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/bitdepth_conversion_sse2.h")
if(NOT CONFIG_AV1_HIGHBITDEPTH)
list(REMOVE_ITEM AOM_DSP_COMMON_INTRIN_SSE2
"${AOM_ROOT}/aom_dsp/x86/highbd_convolve_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_loopfilter_sse2.c")
endif()
list(APPEND AOM_DSP_COMMON_ASM_SSSE3
"${AOM_ROOT}/aom_dsp/x86/aom_subpixel_8t_ssse3.asm"
"${AOM_ROOT}/aom_dsp/x86/aom_subpixel_bilinear_ssse3.asm")
@ -85,6 +91,11 @@ list(APPEND AOM_DSP_COMMON_INTRIN_SSSE3
"${AOM_ROOT}/aom_dsp/x86/highbd_convolve_ssse3.c"
"${AOM_ROOT}/aom_dsp/x86/intrapred_ssse3.c")
if(NOT CONFIG_AV1_HIGHBITDEPTH)
list(REMOVE_ITEM AOM_DSP_COMMON_INTRIN_SSSE3
"${AOM_ROOT}/aom_dsp/x86/highbd_convolve_ssse3.c")
endif()
list(APPEND AOM_DSP_COMMON_INTRIN_SSE4_1
"${AOM_ROOT}/aom_dsp/x86/blend_mask_sse4.h"
"${AOM_ROOT}/aom_dsp/x86/blend_a64_hmask_sse4.c"
@ -104,8 +115,13 @@ list(APPEND AOM_DSP_COMMON_INTRIN_AVX2
"${AOM_ROOT}/aom_dsp/x86/avg_intrin_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/bitdepth_conversion_avx2.h")
list(APPEND AOM_DSP_COMMON_INTRIN_NEON
"${AOM_ROOT}/aom_dsp/arm/fwd_txfm_neon.c"
if(NOT CONFIG_AV1_HIGHBITDEPTH)
list(REMOVE_ITEM AOM_DSP_COMMON_INTRIN_AVX2
"${AOM_ROOT}/aom_dsp/x86/highbd_convolve_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_loopfilter_avx2.c")
endif()
list(APPEND AOM_DSP_COMMON_INTRIN_NEON "${AOM_ROOT}/aom_dsp/arm/fwd_txfm_neon.c"
"${AOM_ROOT}/aom_dsp/arm/loopfilter_neon.c"
"${AOM_ROOT}/aom_dsp/arm/intrapred_neon.c"
"${AOM_ROOT}/aom_dsp/arm/subtract_neon.c"
@ -151,6 +167,7 @@ if(CONFIG_AV1_ENCODER)
"${AOM_ROOT}/aom_dsp/binary_codes_writer.c"
"${AOM_ROOT}/aom_dsp/binary_codes_writer.h"
"${AOM_ROOT}/aom_dsp/bitwriter.h"
"${AOM_ROOT}/aom_dsp/blk_sse_sum.c"
"${AOM_ROOT}/aom_dsp/daalaboolwriter.c"
"${AOM_ROOT}/aom_dsp/daalaboolwriter.h"
"${AOM_ROOT}/aom_dsp/entenc.c"
@ -195,9 +212,18 @@ if(CONFIG_AV1_ENCODER)
"${AOM_ROOT}/aom_dsp/x86/highbd_variance_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/quantize_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/adaptive_quantize_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_adaptive_quantize_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/quantize_x86.h"
"${AOM_ROOT}/aom_dsp/x86/blk_sse_sum_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/sum_squares_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/variance_sse2.c")
if(NOT CONFIG_AV1_HIGHBITDEPTH)
list(REMOVE_ITEM AOM_DSP_ENCODER_INTRIN_SSE2
"${AOM_ROOT}/aom_dsp/x86/highbd_adaptive_quantize_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_quantize_intrin_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_subtract_sse2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_variance_sse2.c")
endif()
list(APPEND AOM_DSP_ENCODER_ASM_SSSE3_X86_64
"${AOM_ROOT}/aom_dsp/x86/fwd_txfm_ssse3_x86_64.asm"
@ -208,6 +234,7 @@ if(CONFIG_AV1_ENCODER)
"${AOM_ROOT}/aom_dsp/x86/subtract_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_quantize_intrin_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/adaptive_quantize_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/highbd_adaptive_quantize_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/sad4d_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/sad_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/sad_highbd_avx2.c"
@ -218,6 +245,7 @@ if(CONFIG_AV1_ENCODER)
"${AOM_ROOT}/aom_dsp/x86/variance_impl_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/obmc_sad_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/obmc_variance_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/blk_sse_sum_avx2.c"
"${AOM_ROOT}/aom_dsp/x86/sum_squares_avx2.c")
list(APPEND AOM_DSP_ENCODER_AVX_ASM_X86_64
@ -239,11 +267,17 @@ if(CONFIG_AV1_ENCODER)
"${AOM_ROOT}/aom_dsp/x86/obmc_sad_sse4.c"
"${AOM_ROOT}/aom_dsp/x86/obmc_variance_sse4.c")
list(APPEND AOM_DSP_ENCODER_INTRIN_NEON
"${AOM_ROOT}/aom_dsp/arm/sad4d_neon.c"
if(NOT CONFIG_AV1_HIGHBITDEPTH)
list(REMOVE_ITEM AOM_DSP_ENCODER_INTRIN_SSE4_1
"${AOM_ROOT}/aom_dsp/x86/highbd_variance_sse4.c")
endif()
list(APPEND AOM_DSP_ENCODER_INTRIN_NEON "${AOM_ROOT}/aom_dsp/arm/sad4d_neon.c"
"${AOM_ROOT}/aom_dsp/arm/sad_neon.c"
"${AOM_ROOT}/aom_dsp/arm/subpel_variance_neon.c"
"${AOM_ROOT}/aom_dsp/arm/variance_neon.c")
"${AOM_ROOT}/aom_dsp/arm/variance_neon.c"
"${AOM_ROOT}/aom_dsp/arm/hadamard_neon.c"
"${AOM_ROOT}/aom_dsp/arm/avg_neon.c")
list(APPEND AOM_DSP_ENCODER_INTRIN_MSA "${AOM_ROOT}/aom_dsp/mips/sad_msa.c"
"${AOM_ROOT}/aom_dsp/mips/subtract_msa.c"

752
third_party/aom/aom_dsp/aom_dsp_rtcd_defs.pl vendored
View file

File diff suppressed because it is too large Load diff

49
third_party/aom/aom_dsp/arm/avg_neon.c vendored Normal file
View file

@ -0,0 +1,49 @@
/*
* Copyright (c) 2019, Alliance for Open Media. 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 <arm_neon.h>
#include "config/aom_dsp_rtcd.h"
#include "aom/aom_integer.h"
#include "aom_dsp/arm/sum_neon.h"
#include "av1/common/arm/mem_neon.h"
#include "av1/common/arm/transpose_neon.h"
unsigned int aom_avg_4x4_neon(const uint8_t *a, int a_stride) {
const uint8x16_t b = load_unaligned_u8q(a, a_stride);
const uint16x8_t c = vaddl_u8(vget_low_u8(b), vget_high_u8(b));
#if defined(__aarch64__)
const uint32_t d = vaddlvq_u16(c);
return (d + 8) >> 4;
#else
const uint32x2_t d = horizontal_add_u16x8(c);
return vget_lane_u32(vrshr_n_u32(d, 4), 0);
#endif
}
unsigned int aom_avg_8x8_neon(const uint8_t *a, int a_stride) {
uint16x8_t sum;
uint32x2_t d;
uint8x8_t b = vld1_u8(a);
a += a_stride;
uint8x8_t c = vld1_u8(a);
a += a_stride;
sum = vaddl_u8(b, c);
for (int i = 0; i < 6; ++i) {
const uint8x8_t e = vld1_u8(a);
a += a_stride;
sum = vaddw_u8(sum, e);
}
d = horizontal_add_u16x8(sum);
return vget_lane_u32(vrshr_n_u32(d, 6), 0);
}

113
third_party/aom/aom_dsp/arm/hadamard_neon.c vendored Normal file
View file

@ -0,0 +1,113 @@
/*
* Copyright (c) 2019, Alliance for Open Media. 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 <arm_neon.h>
#include "config/aom_dsp_rtcd.h"
#include "aom/aom_integer.h"
#include "av1/common/arm/mem_neon.h"
#include "av1/common/arm/transpose_neon.h"
static void hadamard8x8_one_pass(int16x8_t *a0, int16x8_t *a1, int16x8_t *a2,
int16x8_t *a3, int16x8_t *a4, int16x8_t *a5,
int16x8_t *a6, int16x8_t *a7) {
const int16x8_t b0 = vaddq_s16(*a0, *a1);
const int16x8_t b1 = vsubq_s16(*a0, *a1);
const int16x8_t b2 = vaddq_s16(*a2, *a3);
const int16x8_t b3 = vsubq_s16(*a2, *a3);
const int16x8_t b4 = vaddq_s16(*a4, *a5);
const int16x8_t b5 = vsubq_s16(*a4, *a5);
const int16x8_t b6 = vaddq_s16(*a6, *a7);
const int16x8_t b7 = vsubq_s16(*a6, *a7);
const int16x8_t c0 = vaddq_s16(b0, b2);
const int16x8_t c1 = vaddq_s16(b1, b3);
const int16x8_t c2 = vsubq_s16(b0, b2);
const int16x8_t c3 = vsubq_s16(b1, b3);
const int16x8_t c4 = vaddq_s16(b4, b6);
const int16x8_t c5 = vaddq_s16(b5, b7);
const int16x8_t c6 = vsubq_s16(b4, b6);
const int16x8_t c7 = vsubq_s16(b5, b7);
*a0 = vaddq_s16(c0, c4);
*a1 = vsubq_s16(c2, c6);
*a2 = vsubq_s16(c0, c4);
*a3 = vaddq_s16(c2, c6);
*a4 = vaddq_s16(c3, c7);
*a5 = vsubq_s16(c3, c7);
*a6 = vsubq_s16(c1, c5);
*a7 = vaddq_s16(c1, c5);
}
void aom_hadamard_8x8_neon(const int16_t *src_diff, ptrdiff_t src_stride,
tran_low_t *coeff) {
int16x8_t a0 = vld1q_s16(src_diff);
int16x8_t a1 = vld1q_s16(src_diff + src_stride);
int16x8_t a2 = vld1q_s16(src_diff + 2 * src_stride);
int16x8_t a3 = vld1q_s16(src_diff + 3 * src_stride);
int16x8_t a4 = vld1q_s16(src_diff + 4 * src_stride);
int16x8_t a5 = vld1q_s16(src_diff + 5 * src_stride);
int16x8_t a6 = vld1q_s16(src_diff + 6 * src_stride);
int16x8_t a7 = vld1q_s16(src_diff + 7 * src_stride);
hadamard8x8_one_pass(&a0, &a1, &a2, &a3, &a4, &a5, &a6, &a7);
transpose_s16_8x8(&a0, &a1, &a2, &a3, &a4, &a5, &a6, &a7);
hadamard8x8_one_pass(&a0, &a1, &a2, &a3, &a4, &a5, &a6, &a7);
// Skip the second transpose because it is not required.
store_s16q_to_tran_low(coeff + 0, a0);
store_s16q_to_tran_low(coeff + 8, a1);
store_s16q_to_tran_low(coeff + 16, a2);
store_s16q_to_tran_low(coeff + 24, a3);
store_s16q_to_tran_low(coeff + 32, a4);
store_s16q_to_tran_low(coeff + 40, a5);
store_s16q_to_tran_low(coeff + 48, a6);
store_s16q_to_tran_low(coeff + 56, a7);
}
void aom_hadamard_16x16_neon(const int16_t *src_diff, ptrdiff_t src_stride,
tran_low_t *coeff) {
/* Rearrange 16x16 to 8x32 and remove stride.
* Top left first. */
aom_hadamard_8x8_neon(src_diff + 0 + 0 * src_stride, src_stride, coeff + 0);
/* Top right. */
aom_hadamard_8x8_neon(src_diff + 8 + 0 * src_stride, src_stride, coeff + 64);
/* Bottom left. */
aom_hadamard_8x8_neon(src_diff + 0 + 8 * src_stride, src_stride, coeff + 128);
/* Bottom right. */
aom_hadamard_8x8_neon(src_diff + 8 + 8 * src_stride, src_stride, coeff + 192);
for (int i = 0; i < 64; i += 8) {
const int16x8_t a0 = load_tran_low_to_s16q(coeff + 0);
const int16x8_t a1 = load_tran_low_to_s16q(coeff + 64);
const int16x8_t a2 = load_tran_low_to_s16q(coeff + 128);
const int16x8_t a3 = load_tran_low_to_s16q(coeff + 192);
const int16x8_t b0 = vhaddq_s16(a0, a1);
const int16x8_t b1 = vhsubq_s16(a0, a1);
const int16x8_t b2 = vhaddq_s16(a2, a3);
const int16x8_t b3 = vhsubq_s16(a2, a3);
const int16x8_t c0 = vaddq_s16(b0, b2);
const int16x8_t c1 = vaddq_s16(b1, b3);
const int16x8_t c2 = vsubq_s16(b0, b2);
const int16x8_t c3 = vsubq_s16(b1, b3);
store_s16q_to_tran_low(coeff + 0, c0);
store_s16q_to_tran_low(coeff + 64, c1);
store_s16q_to_tran_low(coeff + 128, c2);
store_s16q_to_tran_low(coeff + 192, c3);
coeff += 8;
}
}

37
third_party/aom/aom_dsp/arm/sum_neon.h vendored Normal file
View file

@ -0,0 +1,37 @@
/*
* Copyright (c) 2019, Alliance for Open Media. 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 "config/aom_dsp_rtcd.h"
#include "config/aom_config.h"
#include "aom/aom_integer.h"
#include "aom_ports/mem.h"
static INLINE int horizontal_add_s16x8(const int16x8_t v_16x8) {
const int32x4_t a = vpaddlq_s16(v_16x8);
const int64x2_t b = vpaddlq_s32(a);
const int32x2_t c = vadd_s32(vreinterpret_s32_s64(vget_low_s64(b)),
vreinterpret_s32_s64(vget_high_s64(b)));
return vget_lane_s32(c, 0);
}
static INLINE int horizontal_add_s32x4(const int32x4_t v_32x4) {
const int64x2_t b = vpaddlq_s32(v_32x4);
const int32x2_t c = vadd_s32(vreinterpret_s32_s64(vget_low_s64(b)),
vreinterpret_s32_s64(vget_high_s64(b)));
return vget_lane_s32(c, 0);
}
static INLINE uint32x2_t horizontal_add_u16x8(const uint16x8_t a) {
const uint32x4_t b = vpaddlq_u16(a);
const uint64x2_t c = vpaddlq_u32(b);
return vadd_u32(vreinterpret_u32_u64(vget_low_u64(c)),
vreinterpret_u32_u64(vget_high_u64(c)));
}

17
third_party/aom/aom_dsp/arm/variance_neon.c vendored
View file

@ -13,25 +13,10 @@
#include "config/aom_dsp_rtcd.h"
#include "config/aom_config.h"
#include "aom_dsp/arm/sum_neon.h"
#include "aom/aom_integer.h"
#include "aom_ports/mem.h"
static INLINE int horizontal_add_s16x8(const int16x8_t v_16x8) {
const int32x4_t a = vpaddlq_s16(v_16x8);
const int64x2_t b = vpaddlq_s32(a);
const int32x2_t c = vadd_s32(vreinterpret_s32_s64(vget_low_s64(b)),
vreinterpret_s32_s64(vget_high_s64(b)));
return vget_lane_s32(c, 0);
}
static INLINE int horizontal_add_s32x4(const int32x4_t v_32x4) {
const int64x2_t b = vpaddlq_s32(v_32x4);
const int32x2_t c = vadd_s32(vreinterpret_s32_s64(vget_low_s64(b)),
vreinterpret_s32_s64(vget_high_s64(b)));
return vget_lane_s32(c, 0);
}
// w * h must be less than 2048 or local variable v_sum may overflow.
static void variance_neon_w8(const uint8_t *a, int a_stride, const uint8_t *b,
int b_stride, int w, int h, uint32_t *sse,

198
third_party/aom/aom_dsp/avg.c vendored
View file

@ -48,6 +48,46 @@ unsigned int aom_avg_8x8_c(const uint8_t *s, int p) {
return (sum + 32) >> 6;
}
#if CONFIG_AV1_HIGHBITDEPTH
unsigned int aom_highbd_avg_8x8_c(const uint8_t *s8, int p) {
int i, j;
int sum = 0;
const uint16_t *s = CONVERT_TO_SHORTPTR(s8);
for (i = 0; i < 8; ++i, s += p)
for (j = 0; j < 8; sum += s[j], ++j) {
}
return (sum + 32) >> 6;
}
unsigned int aom_highbd_avg_4x4_c(const uint8_t *s8, int p) {
int i, j;
int sum = 0;
const uint16_t *s = CONVERT_TO_SHORTPTR(s8);
for (i = 0; i < 4; ++i, s += p)
for (j = 0; j < 4; sum += s[j], ++j) {
}
return (sum + 8) >> 4;
}
void aom_highbd_minmax_8x8_c(const uint8_t *s8, int p, const uint8_t *d8,
int dp, int *min, int *max) {
int i, j;
const uint16_t *s = CONVERT_TO_SHORTPTR(s8);
const uint16_t *d = CONVERT_TO_SHORTPTR(d8);
*min = 255;
*max = 0;
for (i = 0; i < 8; ++i, s += p, d += dp) {
for (j = 0; j < 8; ++j) {
int diff = abs(s[j] - d[j]);
*min = diff < *min ? diff : *min;
*max = diff > *max ? diff : *max;
}
}
}
#endif // CONFIG_VP9_HIGHBITDEPTH
// src_diff: first pass, 9 bit, dynamic range [-255, 255]
// second pass, 12 bit, dynamic range [-2040, 2040]
static void hadamard_col8(const int16_t *src_diff, ptrdiff_t src_stride,
@ -170,6 +210,164 @@ void aom_hadamard_32x32_c(const int16_t *src_diff, ptrdiff_t src_stride,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
static void hadamard_highbd_col8_first_pass(const int16_t *src_diff,
ptrdiff_t src_stride,
int16_t *coeff) {
int16_t b0 = src_diff[0 * src_stride] + src_diff[1 * src_stride];
int16_t b1 = src_diff[0 * src_stride] - src_diff[1 * src_stride];
int16_t b2 = src_diff[2 * src_stride] + src_diff[3 * src_stride];
int16_t b3 = src_diff[2 * src_stride] - src_diff[3 * src_stride];
int16_t b4 = src_diff[4 * src_stride] + src_diff[5 * src_stride];
int16_t b5 = src_diff[4 * src_stride] - src_diff[5 * src_stride];
int16_t b6 = src_diff[6 * src_stride] + src_diff[7 * src_stride];
int16_t b7 = src_diff[6 * src_stride] - src_diff[7 * src_stride];
int16_t c0 = b0 + b2;
int16_t c1 = b1 + b3;
int16_t c2 = b0 - b2;
int16_t c3 = b1 - b3;
int16_t c4 = b4 + b6;
int16_t c5 = b5 + b7;
int16_t c6 = b4 - b6;
int16_t c7 = b5 - b7;
coeff[0] = c0 + c4;
coeff[7] = c1 + c5;
coeff[3] = c2 + c6;
coeff[4] = c3 + c7;
coeff[2] = c0 - c4;
coeff[6] = c1 - c5;
coeff[1] = c2 - c6;
coeff[5] = c3 - c7;
}
// src_diff: 16 bit, dynamic range [-32760, 32760]
// coeff: 19 bit
static void hadamard_highbd_col8_second_pass(const int16_t *src_diff,
ptrdiff_t src_stride,
int32_t *coeff) {
int32_t b0 = src_diff[0 * src_stride] + src_diff[1 * src_stride];
int32_t b1 = src_diff[0 * src_stride] - src_diff[1 * src_stride];
int32_t b2 = src_diff[2 * src_stride] + src_diff[3 * src_stride];
int32_t b3 = src_diff[2 * src_stride] - src_diff[3 * src_stride];
int32_t b4 = src_diff[4 * src_stride] + src_diff[5 * src_stride];
int32_t b5 = src_diff[4 * src_stride] - src_diff[5 * src_stride];
int32_t b6 = src_diff[6 * src_stride] + src_diff[7 * src_stride];
int32_t b7 = src_diff[6 * src_stride] - src_diff[7 * src_stride];
int32_t c0 = b0 + b2;
int32_t c1 = b1 + b3;
int32_t c2 = b0 - b2;
int32_t c3 = b1 - b3;
int32_t c4 = b4 + b6;
int32_t c5 = b5 + b7;
int32_t c6 = b4 - b6;
int32_t c7 = b5 - b7;
coeff[0] = c0 + c4;
coeff[7] = c1 + c5;
coeff[3] = c2 + c6;
coeff[4] = c3 + c7;
coeff[2] = c0 - c4;
coeff[6] = c1 - c5;
coeff[1] = c2 - c6;
coeff[5] = c3 - c7;
}
// The order of the output coeff of the hadamard is not important. For
// optimization purposes the final transpose may be skipped.
void aom_highbd_hadamard_8x8_c(const int16_t *src_diff, ptrdiff_t src_stride,
tran_low_t *coeff) {
int idx;
int16_t buffer[64];
int32_t buffer2[64];
int16_t *tmp_buf = &buffer[0];
for (idx = 0; idx < 8; ++idx) {
// src_diff: 13 bit
// buffer: 16 bit, dynamic range [-32760, 32760]
hadamard_highbd_col8_first_pass(src_diff, src_stride, tmp_buf);
tmp_buf += 8;
++src_diff;
}
tmp_buf = &buffer[0];
for (idx = 0; idx < 8; ++idx) {
// buffer: 16 bit
// buffer2: 19 bit, dynamic range [-262080, 262080]
hadamard_highbd_col8_second_pass(tmp_buf, 8, buffer2 + 8 * idx);
++tmp_buf;
}
for (idx = 0; idx < 64; ++idx) coeff[idx] = (tran_low_t)buffer2[idx];
}
// In place 16x16 2D Hadamard transform
void aom_highbd_hadamard_16x16_c(const int16_t *src_diff, ptrdiff_t src_stride,
tran_low_t *coeff) {
int idx;
for (idx = 0; idx < 4; ++idx) {
// src_diff: 13 bit, dynamic range [-4095, 4095]
const int16_t *src_ptr =
src_diff + (idx >> 1) * 8 * src_stride + (idx & 0x01) * 8;
aom_highbd_hadamard_8x8_c(src_ptr, src_stride, coeff + idx * 64);
}
// coeff: 19 bit, dynamic range [-262080, 262080]
for (idx = 0; idx < 64; ++idx) {
tran_low_t a0 = coeff[0];
tran_low_t a1 = coeff[64];
tran_low_t a2 = coeff[128];
tran_low_t a3 = coeff[192];
tran_low_t b0 = (a0 + a1) >> 1;
tran_low_t b1 = (a0 - a1) >> 1;
tran_low_t b2 = (a2 + a3) >> 1;
tran_low_t b3 = (a2 - a3) >> 1;
// new coeff dynamic range: 20 bit
coeff[0] = b0 + b2;
coeff[64] = b1 + b3;
coeff[128] = b0 - b2;
coeff[192] = b1 - b3;
++coeff;
}
}
void aom_highbd_hadamard_32x32_c(const int16_t *src_diff, ptrdiff_t src_stride,
tran_low_t *coeff) {
int idx;
for (idx = 0; idx < 4; ++idx) {
// src_diff: 13 bit, dynamic range [-4095, 4095]
const int16_t *src_ptr =
src_diff + (idx >> 1) * 16 * src_stride + (idx & 0x01) * 16;
aom_highbd_hadamard_16x16_c(src_ptr, src_stride, coeff + idx * 256);
}
// coeff: 20 bit
for (idx = 0; idx < 256; ++idx) {
tran_low_t a0 = coeff[0];
tran_low_t a1 = coeff[256];
tran_low_t a2 = coeff[512];
tran_low_t a3 = coeff[768];
tran_low_t b0 = (a0 + a1) >> 2;
tran_low_t b1 = (a0 - a1) >> 2;
tran_low_t b2 = (a2 + a3) >> 2;
tran_low_t b3 = (a2 - a3) >> 2;
// new coeff dynamic range: 20 bit
coeff[0] = b0 + b2;
coeff[256] = b1 + b3;
coeff[512] = b0 - b2;
coeff[768] = b1 - b3;
++coeff;
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH
// coeff: 16 bits, dynamic range [-32640, 32640].
// length: value range {16, 64, 256, 1024}.
int aom_satd_c(const tran_low_t *coeff, int length) {

21
third_party/aom/aom_dsp/bitwriter_buffer.c vendored
View file

@ -88,8 +88,8 @@ void aom_wb_write_uvlc(struct aom_write_bit_buffer *wb, uint32_t v) {
aom_wb_write_unsigned_literal(wb, v, (leading_zeroes + 1) >> 1);
}
static void aom_wb_write_primitive_quniform(struct aom_write_bit_buffer *wb,
uint16_t n, uint16_t v) {
static void wb_write_primitive_quniform(struct aom_write_bit_buffer *wb,
uint16_t n, uint16_t v) {
if (n <= 1) return;
const int l = get_msb(n) + 1;
const int m = (1 << l) - n;
@ -101,16 +101,15 @@ static void aom_wb_write_primitive_quniform(struct aom_write_bit_buffer *wb,
}
}
static void aom_wb_write_primitive_subexpfin(struct aom_write_bit_buffer *wb,
uint16_t n, uint16_t k,
uint16_t v) {
static void wb_write_primitive_subexpfin(struct aom_write_bit_buffer *wb,
uint16_t n, uint16_t k, uint16_t v) {
int i = 0;
int mk = 0;
while (1) {
int b = (i ? k + i - 1 : k);
int a = (1 << b);
if (n <= mk + 3 * a) {
aom_wb_write_primitive_quniform(wb, n - mk, v - mk);
wb_write_primitive_quniform(wb, n - mk, v - mk);
break;
} else {
int t = (v >= mk + a);
@ -126,10 +125,10 @@ static void aom_wb_write_primitive_subexpfin(struct aom_write_bit_buffer *wb,
}
}
static void aom_wb_write_primitive_refsubexpfin(struct aom_write_bit_buffer *wb,
uint16_t n, uint16_t k,
uint16_t ref, uint16_t v) {
aom_wb_write_primitive_subexpfin(wb, n, k, recenter_finite_nonneg(n, ref, v));
static void wb_write_primitive_refsubexpfin(struct aom_write_bit_buffer *wb,
uint16_t n, uint16_t k,
uint16_t ref, uint16_t v) {
wb_write_primitive_subexpfin(wb, n, k, recenter_finite_nonneg(n, ref, v));
}
void aom_wb_write_signed_primitive_refsubexpfin(struct aom_write_bit_buffer *wb,
@ -138,5 +137,5 @@ void aom_wb_write_signed_primitive_refsubexpfin(struct aom_write_bit_buffer *wb,
ref += n - 1;
v += n - 1;
const uint16_t scaled_n = (n << 1) - 1;
aom_wb_write_primitive_refsubexpfin(wb, scaled_n, k, ref, v);
wb_write_primitive_refsubexpfin(wb, scaled_n, k, ref, v);
}

2
third_party/aom/aom_dsp/blend_a64_hmask.c vendored
View file

@ -40,6 +40,7 @@ void aom_blend_a64_hmask_c(uint8_t *dst, uint32_t dst_stride,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_blend_a64_hmask_c(uint8_t *dst_8, uint32_t dst_stride,
const uint8_t *src0_8, uint32_t src0_stride,
const uint8_t *src1_8, uint32_t src1_stride,
@ -67,3 +68,4 @@ void aom_highbd_blend_a64_hmask_c(uint8_t *dst_8, uint32_t dst_stride,
}
}
}
#endif

4
third_party/aom/aom_dsp/blend_a64_mask.c vendored
View file

@ -120,6 +120,7 @@ void aom_lowbd_blend_a64_d16_mask_c(
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_blend_a64_d16_mask_c(
uint8_t *dst_8, uint32_t dst_stride, const CONV_BUF_TYPE *src0,
uint32_t src0_stride, const CONV_BUF_TYPE *src1, uint32_t src1_stride,
@ -219,6 +220,7 @@ void aom_highbd_blend_a64_d16_mask_c(
}
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH
// Blending with alpha mask. Mask values come from the range [0, 64],
// as described for AOM_BLEND_A64 in aom_dsp/blend.h. src0 or src1 can
@ -281,6 +283,7 @@ void aom_blend_a64_mask_c(uint8_t *dst, uint32_t dst_stride,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_blend_a64_mask_c(uint8_t *dst_8, uint32_t dst_stride,
const uint8_t *src0_8, uint32_t src0_stride,
const uint8_t *src1_8, uint32_t src1_stride,
@ -343,3 +346,4 @@ void aom_highbd_blend_a64_mask_c(uint8_t *dst_8, uint32_t dst_stride,
}
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH

2
third_party/aom/aom_dsp/blend_a64_vmask.c vendored
View file

@ -41,6 +41,7 @@ void aom_blend_a64_vmask_c(uint8_t *dst, uint32_t dst_stride,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_blend_a64_vmask_c(uint8_t *dst_8, uint32_t dst_stride,
const uint8_t *src0_8, uint32_t src0_stride,
const uint8_t *src1_8, uint32_t src1_stride,
@ -69,3 +70,4 @@ void aom_highbd_blend_a64_vmask_c(uint8_t *dst_8, uint32_t dst_stride,
}
}
}
#endif

26
third_party/aom/aom_dsp/blk_sse_sum.c vendored Normal file
View file

@ -0,0 +1,26 @@
/*
* Copyright (c) 2019, Alliance for Open Media. All rights reserved
*
* This source code is subject to the terms of the BSD 2 Clause License and
* the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
* was not distributed with this source code in the LICENSE file, you can
* obtain it at www.aomedia.org/license/software. If the Alliance for Open
* Media Patent License 1.0 was not distributed with this source code in the
* PATENTS file, you can obtain it at www.aomedia.org/license/patent.
*/
#include "config/aom_dsp_rtcd.h"
void aom_get_blk_sse_sum_c(const int16_t *data, int stride, int bw, int bh,
int *x_sum, int64_t *x2_sum) {
*x_sum = 0;
*x2_sum = 0;
for (int i = 0; i < bh; ++i) {
for (int j = 0; j < bw; ++j) {
const int val = data[j];
*x_sum += val;
*x2_sum += val * val;
}
data += stride;
}
}

6
third_party/aom/aom_dsp/daalaboolreader.h vendored
View file

@ -54,7 +54,7 @@ static INLINE int aom_daala_read(daala_reader *r, int prob) {
#if CONFIG_BITSTREAM_DEBUG
/*{
const int queue_r = bitstream_queue_get_read();
const int frame_idx = bitstream_queue_get_frame_read();
const int frame_idx = aom_bitstream_queue_get_frame_read();
if (frame_idx == 0 && queue_r == 0) {
fprintf(stderr, "\n *** bitstream queue at frame_idx_r %d queue_r %d\n",
frame_idx, queue_r);
@ -70,7 +70,7 @@ static INLINE int aom_daala_read(daala_reader *r, int prob) {
int ref_bit, ref_nsymbs;
aom_cdf_prob ref_cdf[16];
const int queue_r = bitstream_queue_get_read();
const int frame_idx = bitstream_queue_get_frame_read();
const int frame_idx = aom_bitstream_queue_get_frame_read();
bitstream_queue_pop(&ref_bit, ref_cdf, &ref_nsymbs);
if (ref_nsymbs != 2) {
fprintf(stderr,
@ -114,7 +114,7 @@ static INLINE int daala_read_symbol(daala_reader *r, const aom_cdf_prob *cdf,
int ref_symb, ref_nsymbs;
aom_cdf_prob ref_cdf[16];
const int queue_r = bitstream_queue_get_read();
const int frame_idx = bitstream_queue_get_frame_read();
const int frame_idx = aom_bitstream_queue_get_frame_read();
bitstream_queue_pop(&ref_symb, ref_cdf, &ref_nsymbs);
if (nsymbs != ref_nsymbs) {
fprintf(stderr,

4
third_party/aom/aom_dsp/daalaboolwriter.h vendored
View file

@ -43,7 +43,7 @@ static INLINE void aom_daala_write(daala_writer *w, int bit, int prob) {
/*int queue_r = 0;
int frame_idx_r = 0;
int queue_w = bitstream_queue_get_write();
int frame_idx_w = bitstream_queue_get_frame_write();
int frame_idx_w = aom_bitstream_queue_get_frame_writee();
if (frame_idx_w == frame_idx_r && queue_w == queue_r) {
fprintf(stderr, "\n *** bitstream queue at frame_idx_w %d queue_w %d\n",
frame_idx_w, queue_w);
@ -60,7 +60,7 @@ static INLINE void daala_write_symbol(daala_writer *w, int symb,
/*int queue_r = 0;
int frame_idx_r = 0;
int queue_w = bitstream_queue_get_write();
int frame_idx_w = bitstream_queue_get_frame_write();
int frame_idx_w = aom_bitstream_queue_get_frame_writee();
if (frame_idx_w == frame_idx_r && queue_w == queue_r) {
fprintf(stderr, "\n *** bitstream queue at frame_idx_w %d queue_w %d\n",
frame_idx_w, queue_w);

2
third_party/aom/aom_dsp/fwd_txfm.c vendored
View file

@ -97,7 +97,9 @@ void aom_fdct8x8_c(const int16_t *input, tran_low_t *final_output, int stride) {
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_fdct8x8_c(const int16_t *input, tran_low_t *final_output,
int stride) {
aom_fdct8x8_c(input, final_output, stride);
}
#endif

2
third_party/aom/aom_dsp/grain_table.c vendored
View file

@ -203,7 +203,7 @@ int aom_film_grain_table_lookup(aom_film_grain_table_t *t, int64_t time_stamp,
aom_film_grain_t *grain) {
aom_film_grain_table_entry_t *entry = t->head;
aom_film_grain_table_entry_t *prev_entry = 0;
int16_t random_seed = grain ? grain->random_seed : 0;
uint16_t random_seed = grain ? grain->random_seed : 0;
if (grain) memset(grain, 0, sizeof(*grain));
while (entry) {

32
third_party/aom/aom_dsp/loopfilter.c vendored
View file

@ -21,6 +21,7 @@ static INLINE int8_t signed_char_clamp(int t) {
return (int8_t)clamp(t, -128, 127);
}
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE int16_t signed_char_clamp_high(int t, int bd) {
switch (bd) {
case 10: return (int16_t)clamp(t, -128 * 4, 128 * 4 - 1);
@ -29,6 +30,7 @@ static INLINE int16_t signed_char_clamp_high(int t, int bd) {
default: return (int16_t)clamp(t, -128, 128 - 1);
}
}
#endif
// should we apply any filter at all: 11111111 yes, 00000000 no
static INLINE int8_t filter_mask2(uint8_t limit, uint8_t blimit, uint8_t p1,
@ -103,11 +105,11 @@ static INLINE void filter4(int8_t mask, uint8_t thresh, uint8_t *op1,
uint8_t *op0, uint8_t *oq0, uint8_t *oq1) {
int8_t filter1, filter2;
const int8_t ps1 = (int8_t)*op1 ^ 0x80;
const int8_t ps0 = (int8_t)*op0 ^ 0x80;
const int8_t qs0 = (int8_t)*oq0 ^ 0x80;
const int8_t qs1 = (int8_t)*oq1 ^ 0x80;
const uint8_t hev = hev_mask(thresh, *op1, *op0, *oq0, *oq1);
const int8_t ps1 = (int8_t)(*op1 ^ 0x80);
const int8_t ps0 = (int8_t)(*op0 ^ 0x80);
const int8_t qs0 = (int8_t)(*oq0 ^ 0x80);
const int8_t qs1 = (int8_t)(*oq1 ^ 0x80);
const int8_t hev = hev_mask(thresh, *op1, *op0, *oq0, *oq1);
// add outer taps if we have high edge variance
int8_t filter = signed_char_clamp(ps1 - qs1) & hev;
@ -121,14 +123,14 @@ static INLINE void filter4(int8_t mask, uint8_t thresh, uint8_t *op1,
filter1 = signed_char_clamp(filter + 4) >> 3;
filter2 = signed_char_clamp(filter + 3) >> 3;
*oq0 = signed_char_clamp(qs0 - filter1) ^ 0x80;
*op0 = signed_char_clamp(ps0 + filter2) ^ 0x80;
*oq0 = (uint8_t)(signed_char_clamp(qs0 - filter1) ^ 0x80);
*op0 = (uint8_t)(signed_char_clamp(ps0 + filter2) ^ 0x80);
// outer tap adjustments
filter = ROUND_POWER_OF_TWO(filter1, 1) & ~hev;
*oq1 = signed_char_clamp(qs1 - filter) ^ 0x80;
*op1 = signed_char_clamp(ps1 + filter) ^ 0x80;
*oq1 = (uint8_t)(signed_char_clamp(qs1 - filter) ^ 0x80);
*op1 = (uint8_t)(signed_char_clamp(ps1 + filter) ^ 0x80);
}
void aom_lpf_horizontal_4_c(uint8_t *s, int p /* pitch */,
@ -442,6 +444,7 @@ void aom_lpf_vertical_14_dual_c(uint8_t *s, int pitch, const uint8_t *blimit0,
mb_lpf_vertical_edge_w(s + 4 * pitch, pitch, blimit1, limit1, thresh1, 4);
}
#if CONFIG_AV1_HIGHBITDEPTH
// Should we apply any filter at all: 11111111 yes, 00000000 no ?
static INLINE int8_t highbd_filter_mask2(uint8_t limit, uint8_t blimit,
uint16_t p1, uint16_t p0, uint16_t q0,
@ -539,7 +542,7 @@ static INLINE void highbd_filter4(int8_t mask, uint8_t thresh, uint16_t *op1,
const int16_t ps0 = (int16_t)*op0 - (0x80 << shift);
const int16_t qs0 = (int16_t)*oq0 - (0x80 << shift);
const int16_t qs1 = (int16_t)*oq1 - (0x80 << shift);
const uint16_t hev = highbd_hev_mask(thresh, *op1, *op0, *oq0, *oq1, bd);
const int16_t hev = highbd_hev_mask(thresh, *op1, *op0, *oq0, *oq1, bd);
// Add outer taps if we have high edge variance.
int16_t filter = signed_char_clamp_high(ps1 - qs1, bd) & hev;
@ -865,10 +868,10 @@ static void highbd_mb_lpf_horizontal_edge_w(uint16_t *s, int p,
}
}
void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int p, const uint8_t *blimit,
const uint8_t *limit, const uint8_t *thresh,
int bd) {
highbd_mb_lpf_horizontal_edge_w(s, p, blimit, limit, thresh, 1, bd);
void aom_highbd_lpf_horizontal_14_c(uint16_t *s, int pitch,
const uint8_t *blimit, const uint8_t *limit,
const uint8_t *thresh, int bd) {
highbd_mb_lpf_horizontal_edge_w(s, pitch, blimit, limit, thresh, 1, bd);
}
void aom_highbd_lpf_horizontal_14_dual_c(
@ -923,3 +926,4 @@ void aom_highbd_lpf_vertical_14_dual_c(
highbd_mb_lpf_vertical_edge_w(s + 4 * pitch, pitch, blimit1, limit1, thresh1,
4, bd);
}
#endif // CONFIG_AV1_HIGHBITDEPTH

4
third_party/aom/aom_dsp/noise_util.c vendored
View file

@ -96,7 +96,9 @@ void aom_noise_tx_filter(struct aom_noise_tx_t *noise_tx, const float *psd) {
for (int x = 0; x < block_size; ++x) {
int i = y * block_size + x;
float *c = noise_tx->tx_block + 2 * i;
const float p = c[0] * c[0] + c[1] * c[1];
const float c0 = AOMMAX((float)fabs(c[0]), 1e-8f);
const float c1 = AOMMAX((float)fabs(c[1]), 1e-8f);
const float p = c0 * c0 + c1 * c1;
if (p > kBeta * psd[i] && p > 1e-6) {
noise_tx->tx_block[2 * i + 0] *= (p - psd[i]) / AOMMAX(p, kEps);
noise_tx->tx_block[2 * i + 1] *= (p - psd[i]) / AOMMAX(p, kEps);

19
third_party/aom/aom_dsp/psnr.c vendored
View file

@ -49,6 +49,7 @@ static void encoder_variance(const uint8_t *a, int a_stride, const uint8_t *b,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
static void encoder_highbd_variance64(const uint8_t *a8, int a_stride,
const uint8_t *b8, int b_stride, int w,
int h, uint64_t *sse, int64_t *sum) {
@ -81,6 +82,7 @@ static void encoder_highbd_8_variance(const uint8_t *a8, int a_stride,
*sse = (unsigned int)sse_long;
*sum = (int)sum_long;
}
#endif // CONFIG_AV1_HIGHBITDEPTH
static int64_t get_sse(const uint8_t *a, int a_stride, const uint8_t *b,
int b_stride, int width, int height) {
@ -122,6 +124,7 @@ static int64_t get_sse(const uint8_t *a, int a_stride, const uint8_t *b,
return total_sse;
}
#if CONFIG_AV1_HIGHBITDEPTH
static int64_t highbd_get_sse_shift(const uint8_t *a8, int a_stride,
const uint8_t *b8, int b_stride, int width,
int height, unsigned int input_shift) {
@ -174,6 +177,7 @@ static int64_t highbd_get_sse(const uint8_t *a, int a_stride, const uint8_t *b,
}
return total_sse;
}
#endif // CONFIG_AV1_HIGHBITDEPTH
int64_t aom_get_y_sse_part(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, int hstart, int width,
@ -226,6 +230,7 @@ int64_t aom_get_v_sse(const YV12_BUFFER_CONFIG *a,
a->uv_crop_width, a->uv_crop_height);
}
#if CONFIG_AV1_HIGHBITDEPTH
int64_t aom_highbd_get_y_sse_part(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, int hstart,
int width, int vstart, int height) {
@ -284,9 +289,11 @@ int64_t aom_highbd_get_v_sse(const YV12_BUFFER_CONFIG *a,
return highbd_get_sse(a->v_buffer, a->uv_stride, b->v_buffer, b->uv_stride,
a->uv_crop_width, a->uv_crop_height);
}
#endif // CONFIG_AV1_HIGHBITDEPTH
int64_t aom_get_sse_plane(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, int plane, int highbd) {
#if CONFIG_AV1_HIGHBITDEPTH
if (highbd) {
switch (plane) {
case 0: return aom_highbd_get_y_sse(a, b);
@ -294,15 +301,26 @@ int64_t aom_get_sse_plane(const YV12_BUFFER_CONFIG *a,
case 2: return aom_highbd_get_v_sse(a, b);
default: assert(plane >= 0 && plane <= 2); return 0;
}
} else {
switch (plane) {
case 0: return aom_get_y_sse(a, b);
case 1: return aom_get_u_sse(a, b);
case 2: return aom_get_v_sse(a, b);
default: assert(plane >= 0 && plane <= 2); return 0;
}
}
#else
(void)highbd;
switch (plane) {
case 0: return aom_get_y_sse(a, b);
case 1: return aom_get_u_sse(a, b);
case 2: return aom_get_v_sse(a, b);
default: assert(plane >= 0 && plane <= 2); return 0;
}
#endif
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_calc_highbd_psnr(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, PSNR_STATS *psnr,
uint32_t bit_depth, uint32_t in_bit_depth) {
@ -347,6 +365,7 @@ void aom_calc_highbd_psnr(const YV12_BUFFER_CONFIG *a,
psnr->psnr[0] =
aom_sse_to_psnr((double)total_samples, peak, (double)total_sse);
}
#endif
void aom_calc_psnr(const YV12_BUFFER_CONFIG *a, const YV12_BUFFER_CONFIG *b,
PSNR_STATS *psnr) {

2
third_party/aom/aom_dsp/psnr.h vendored
View file

@ -49,6 +49,7 @@ int64_t aom_get_v_sse_part(const YV12_BUFFER_CONFIG *a,
int64_t aom_get_v_sse(const YV12_BUFFER_CONFIG *a, const YV12_BUFFER_CONFIG *b);
int64_t aom_get_sse_plane(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, int plane, int highbd);
#if CONFIG_AV1_HIGHBITDEPTH
int64_t aom_highbd_get_y_sse_part(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, int hstart,
int width, int vstart, int height);
@ -67,6 +68,7 @@ int64_t aom_highbd_get_v_sse(const YV12_BUFFER_CONFIG *a,
void aom_calc_highbd_psnr(const YV12_BUFFER_CONFIG *a,
const YV12_BUFFER_CONFIG *b, PSNR_STATS *psnr,
unsigned int bit_depth, unsigned int in_bit_depth);
#endif
void aom_calc_psnr(const YV12_BUFFER_CONFIG *a, const YV12_BUFFER_CONFIG *b,
PSNR_STATS *psnr);

18
third_party/aom/aom_dsp/psnrhvs.c vendored
View file

@ -131,8 +131,24 @@ static double calc_psnrhvs(const unsigned char *src, int _systride,
int pixels;
int x;
int y;
float sum1;
float sum2;
float delt;
(void)_par;
ret = pixels = 0;
sum1 = sum2 = delt = 0.0f;
for (y = 0; y < _h; y++) {
for (x = 0; x < _w; x++) {
if (!buf_is_hbd) {
sum1 += _src8[y * _systride + x];
sum2 += _dst8[y * _dystride + x];
} else {
sum1 += _src16[y * _systride + x] >> _shift;
sum2 += _dst16[y * _dystride + x] >> _shift;
}
}
}
delt = (sum1 - sum2) / (_w * _h);
/*In the PSNR-HVS-M paper[1] the authors describe the construction of
their masking table as "we have used the quantization table for the
color component Y of JPEG [6] that has been also obtained on the
@ -183,6 +199,7 @@ static double calc_psnrhvs(const unsigned char *src, int _systride,
dct_s[i * 8 + j] = _src16[(y + i) * _systride + (j + x)] >> _shift;
dct_d[i * 8 + j] = _dst16[(y + i) * _dystride + (j + x)] >> _shift;
}
dct_d[i * 8 + j] += (int)(delt + 0.5f);
s_gmean += dct_s[i * 8 + j];
d_gmean += dct_d[i * 8 + j];
s_means[sub] += dct_s[i * 8 + j];
@ -242,6 +259,7 @@ static double calc_psnrhvs(const unsigned char *src, int _systride,
}
if (pixels <= 0) return 0;
ret /= pixels;
ret += 0.04 * delt * delt;
return ret;
}

178
third_party/aom/aom_dsp/quantize.c vendored
View file

@ -13,7 +13,7 @@
#include "aom_mem/aom_mem.h"
#include "av1/encoder/av1_quantize.h"
void quantize_b_adaptive_helper_c(
void aom_quantize_b_adaptive_helper_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
@ -104,14 +104,15 @@ void quantize_b_adaptive_helper_c(
*eob_ptr = eob + 1;
}
void quantize_b_helper_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr,
uint16_t *eob_ptr, const int16_t *scan,
const int16_t *iscan, const qm_val_t *qm_ptr,
const qm_val_t *iqm_ptr, const int log_scale) {
void aom_quantize_b_helper_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
const int16_t *quant_ptr,
const int16_t *quant_shift_ptr,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan,
const qm_val_t *qm_ptr, const qm_val_t *iqm_ptr,
const int log_scale) {
const int zbins[2] = { ROUND_POWER_OF_TWO(zbin_ptr[0], log_scale),
ROUND_POWER_OF_TWO(zbin_ptr[1], log_scale) };
const int nzbins[2] = { zbins[0] * -1, zbins[1] * -1 };
@ -166,21 +167,19 @@ void quantize_b_helper_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
*eob_ptr = eob + 1;
}
void highbd_quantize_b_adaptive_helper_c(
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_quantize_b_adaptive_helper_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan, const qm_val_t *qm_ptr,
const qm_val_t *iqm_ptr, const int log_scale) {
int i, eob = -1;
const int zbins[2] = { ROUND_POWER_OF_TWO(zbin_ptr[0], log_scale),
ROUND_POWER_OF_TWO(zbin_ptr[1], log_scale) };
const int nzbins[2] = { zbins[0] * -1, zbins[1] * -1 };
int dequant;
int idx_arr[4096];
(void)iscan;
int idx = 0;
int i, non_zero_count = (int)n_coeffs, eob = -1;
memset(qcoeff_ptr, 0, n_coeffs * sizeof(*qcoeff_ptr));
memset(dqcoeff_ptr, 0, n_coeffs * sizeof(*dqcoeff_ptr));
@ -190,47 +189,49 @@ void highbd_quantize_b_adaptive_helper_c(
prescan_add[i] = ROUND_POWER_OF_TWO(dequant_ptr[i] * EOB_FACTOR, 7);
// Pre-scan pass
for (i = 0; i < n_coeffs; i++) {
for (i = (int)n_coeffs - 1; i >= 0; i--) {
const int rc = scan[i];
const qm_val_t wt = qm_ptr != NULL ? qm_ptr[rc] : (1 << AOM_QM_BITS);
const int coeff = coeff_ptr[rc] * wt;
// If the coefficient is out of the base ZBIN range, keep it for
// quantization.
const int prescan_add_val = prescan_add[rc != 0];
if (coeff >= (zbins[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val) &&
coeff <= (nzbins[rc != 0] * (1 << AOM_QM_BITS) - prescan_add_val))
idx_arr[idx++] = i;
if (coeff < (zbins[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val) &&
coeff > (nzbins[rc != 0] * (1 << AOM_QM_BITS) - prescan_add_val))
non_zero_count--;
else
break;
}
// Quantization pass: only process the coefficients selected in
// pre-scan pass. Note: idx can be zero.
// Quantization pass: All coefficients with index >= zero_flag are
// skippable. Note: zero_flag can be zero.
#if SKIP_EOB_FACTOR_ADJUST
int first = -1;
#endif // SKIP_EOB_FACTOR_ADJUST
for (i = 0; i < idx; i++) {
const int rc = scan[idx_arr[i]];
for (i = 0; i < non_zero_count; i++) {
const int rc = scan[i];
const int coeff = coeff_ptr[rc];
const int coeff_sign = (coeff >> 31);
const qm_val_t wt = qm_ptr != NULL ? qm_ptr[rc] : (1 << AOM_QM_BITS);
const qm_val_t iwt = iqm_ptr != NULL ? iqm_ptr[rc] : (1 << AOM_QM_BITS);
const int abs_coeff = (coeff ^ coeff_sign) - coeff_sign;
const int64_t tmp1 =
abs_coeff + ROUND_POWER_OF_TWO(round_ptr[rc != 0], log_scale);
const int64_t tmpw = tmp1 * wt;
const int64_t tmp2 = ((tmpw * quant_ptr[rc != 0]) >> 16) + tmpw;
const int abs_qcoeff = (int)((tmp2 * quant_shift_ptr[rc != 0]) >>
(16 - log_scale + AOM_QM_BITS));
qcoeff_ptr[rc] = (tran_low_t)((abs_qcoeff ^ coeff_sign) - coeff_sign);
dequant =
(dequant_ptr[rc != 0] * iwt + (1 << (AOM_QM_BITS - 1))) >> AOM_QM_BITS;
const tran_low_t abs_dqcoeff = (abs_qcoeff * dequant) >> log_scale;
dqcoeff_ptr[rc] = (tran_low_t)((abs_dqcoeff ^ coeff_sign) - coeff_sign);
if (abs_qcoeff) {
eob = idx_arr[i];
if (abs_coeff * wt >= (zbins[rc != 0] << AOM_QM_BITS)) {
const int64_t tmp1 =
abs_coeff + ROUND_POWER_OF_TWO(round_ptr[rc != 0], log_scale);
const int64_t tmpw = tmp1 * wt;
const int64_t tmp2 = ((tmpw * quant_ptr[rc != 0]) >> 16) + tmpw;
const int abs_qcoeff = (int)((tmp2 * quant_shift_ptr[rc != 0]) >>
(16 - log_scale + AOM_QM_BITS));
qcoeff_ptr[rc] = (tran_low_t)((abs_qcoeff ^ coeff_sign) - coeff_sign);
const qm_val_t iwt = iqm_ptr != NULL ? iqm_ptr[rc] : (1 << AOM_QM_BITS);
const int dequant =
(dequant_ptr[rc != 0] * iwt + (1 << (AOM_QM_BITS - 1))) >>
AOM_QM_BITS;
const tran_low_t abs_dqcoeff = (abs_qcoeff * dequant) >> log_scale;
dqcoeff_ptr[rc] = (tran_low_t)((abs_dqcoeff ^ coeff_sign) - coeff_sign);
if (abs_qcoeff) {
eob = i;
#if SKIP_EOB_FACTOR_ADJUST
if (first == -1) first = eob;
if (first == -1) first = eob;
#endif // SKIP_EOB_FACTOR_ADJUST
}
}
}
#if SKIP_EOB_FACTOR_ADJUST
@ -254,7 +255,7 @@ void highbd_quantize_b_adaptive_helper_c(
*eob_ptr = eob + 1;
}
void highbd_quantize_b_helper_c(
void aom_highbd_quantize_b_helper_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
@ -310,6 +311,7 @@ void highbd_quantize_b_helper_c(
}
*eob_ptr = eob + 1;
}
#endif // CONFIG_AV1_HIGHBITDEPTH
/* These functions should only be called when quantisation matrices
are not used. */
@ -321,10 +323,10 @@ void aom_quantize_b_adaptive_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 0);
aom_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan,
iscan, NULL, NULL, 0);
}
void aom_quantize_b_32x32_adaptive_c(
@ -333,10 +335,10 @@ void aom_quantize_b_32x32_adaptive_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 1);
aom_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan,
iscan, NULL, NULL, 1);
}
void aom_quantize_b_64x64_adaptive_c(
@ -345,22 +347,23 @@ void aom_quantize_b_64x64_adaptive_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 2);
aom_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan,
iscan, NULL, NULL, 2);
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_quantize_b_adaptive_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
highbd_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan,
iscan, NULL, NULL, 0);
aom_highbd_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr,
round_ptr, quant_ptr, quant_shift_ptr,
qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 0);
}
void aom_highbd_quantize_b_32x32_adaptive_c(
@ -369,10 +372,10 @@ void aom_highbd_quantize_b_32x32_adaptive_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
highbd_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan,
iscan, NULL, NULL, 1);
aom_highbd_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr,
round_ptr, quant_ptr, quant_shift_ptr,
qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 1);
}
void aom_highbd_quantize_b_64x64_adaptive_c(
@ -381,11 +384,12 @@ void aom_highbd_quantize_b_64x64_adaptive_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
highbd_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan,
iscan, NULL, NULL, 2);
aom_highbd_quantize_b_adaptive_helper_c(coeff_ptr, n_coeffs, zbin_ptr,
round_ptr, quant_ptr, quant_shift_ptr,
qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 2);
}
#endif // CONFIG_AV1_HIGHBITDEPTH
void aom_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
@ -393,9 +397,9 @@ void aom_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr, quant_ptr,
quant_shift_ptr, qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 0);
aom_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr, quant_ptr,
quant_shift_ptr, qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 0);
}
void aom_quantize_b_32x32_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
@ -405,9 +409,9 @@ void aom_quantize_b_32x32_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr, quant_ptr,
quant_shift_ptr, qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 1);
aom_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr, quant_ptr,
quant_shift_ptr, qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 1);
}
void aom_quantize_b_64x64_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
@ -417,11 +421,12 @@ void aom_quantize_b_64x64_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr, quant_ptr,
quant_shift_ptr, qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 2);
aom_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr, quant_ptr,
quant_shift_ptr, qcoeff_ptr, dqcoeff_ptr, dequant_ptr,
eob_ptr, scan, iscan, NULL, NULL, 2);
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
const int16_t *quant_ptr,
@ -429,10 +434,10 @@ void aom_highbd_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
highbd_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 0);
aom_highbd_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 0);
}
void aom_highbd_quantize_b_32x32_c(
@ -441,10 +446,10 @@ void aom_highbd_quantize_b_32x32_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
highbd_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 1);
aom_highbd_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 1);
}
void aom_highbd_quantize_b_64x64_c(
@ -453,8 +458,9 @@ void aom_highbd_quantize_b_64x64_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
highbd_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 2);
aom_highbd_quantize_b_helper_c(coeff_ptr, n_coeffs, zbin_ptr, round_ptr,
quant_ptr, quant_shift_ptr, qcoeff_ptr,
dqcoeff_ptr, dequant_ptr, eob_ptr, scan, iscan,
NULL, NULL, 2);
}
#endif // CONFIG_AV1_HIGHBITDEPTH

27
third_party/aom/aom_dsp/quantize.h vendored
View file

@ -20,7 +20,7 @@
extern "C" {
#endif
void quantize_b_adaptive_helper_c(
void aom_quantize_b_adaptive_helper_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
@ -51,7 +51,8 @@ void aom_quantize_b_64x64_adaptive_c(
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan);
void highbd_quantize_b_adaptive_helper_c(
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_quantize_b_adaptive_helper_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
@ -79,15 +80,17 @@ void aom_highbd_quantize_b_64x64_adaptive_c(
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan);
#endif // CONFIG_AV1_HIGHBITDEPTH
void quantize_b_helper_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr,
uint16_t *eob_ptr, const int16_t *scan,
const int16_t *iscan, const qm_val_t *qm_ptr,
const qm_val_t *iqm_ptr, const int log_scale);
void aom_quantize_b_helper_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
const int16_t *quant_ptr,
const int16_t *quant_shift_ptr,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan,
const qm_val_t *qm_ptr, const qm_val_t *iqm_ptr,
const int log_scale);
void aom_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *zbin_ptr, const int16_t *round_ptr,
@ -96,7 +99,8 @@ void aom_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan);
void highbd_quantize_b_helper_c(
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_quantize_b_helper_c(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
@ -111,6 +115,7 @@ void aom_highbd_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs,
tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr,
const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan);
#endif // CONFIG_AV1_HIGHBITDEPTH
#ifdef __cplusplus
} // extern "C"

19
third_party/aom/aom_dsp/sad.c vendored
View file

@ -159,6 +159,7 @@ sadMxNx4D(16, 64);
sadMxN(64, 16);
sadMxNx4D(64, 16);
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE unsigned int highbd_sad(const uint8_t *a8, int a_stride,
const uint8_t *b8, int b_stride,
int width, int height) {
@ -178,11 +179,12 @@ static INLINE unsigned int highbd_sad(const uint8_t *a8, int a_stride,
}
static INLINE unsigned int highbd_sadb(const uint8_t *a8, int a_stride,
const uint16_t *b, int b_stride,
const uint8_t *b8, int b_stride,
int width, int height) {
int y, x;
unsigned int sad = 0;
const uint16_t *a = CONVERT_TO_SHORTPTR(a8);
const uint16_t *b = CONVERT_TO_SHORTPTR(b8);
for (y = 0; y < height; y++) {
for (x = 0; x < width; x++) {
sad += abs(a[x] - b[x]);
@ -204,18 +206,18 @@ static INLINE unsigned int highbd_sadb(const uint8_t *a8, int a_stride,
const uint8_t *src, int src_stride, const uint8_t *ref, int ref_stride, \
const uint8_t *second_pred) { \
uint16_t comp_pred[m * n]; \
aom_highbd_comp_avg_pred(CONVERT_TO_BYTEPTR(comp_pred), second_pred, m, n, \
ref, ref_stride); \
return highbd_sadb(src, src_stride, comp_pred, m, m, n); \
uint8_t *const comp_pred8 = CONVERT_TO_BYTEPTR(comp_pred); \
aom_highbd_comp_avg_pred(comp_pred8, second_pred, m, n, ref, ref_stride); \
return highbd_sadb(src, src_stride, comp_pred8, m, m, n); \
} \
unsigned int aom_highbd_dist_wtd_sad##m##x##n##_avg_c( \
const uint8_t *src, int src_stride, const uint8_t *ref, int ref_stride, \
const uint8_t *second_pred, const DIST_WTD_COMP_PARAMS *jcp_param) { \
uint16_t comp_pred[m * n]; \
aom_highbd_dist_wtd_comp_avg_pred(CONVERT_TO_BYTEPTR(comp_pred), \
second_pred, m, n, ref, ref_stride, \
jcp_param); \
return highbd_sadb(src, src_stride, comp_pred, m, m, n); \
uint8_t *const comp_pred8 = CONVERT_TO_BYTEPTR(comp_pred); \
aom_highbd_dist_wtd_comp_avg_pred(comp_pred8, second_pred, m, n, ref, \
ref_stride, jcp_param); \
return highbd_sadb(src, src_stride, comp_pred8, m, m, n); \
}
#define highbd_sadMxNx4D(m, n) \
@ -305,3 +307,4 @@ highbd_sadMxN(16, 64);
highbd_sadMxNx4D(16, 64);
highbd_sadMxN(64, 16);
highbd_sadMxNx4D(64, 16);
#endif // CONFIG_AV1_HIGHBITDEPTH

13
third_party/aom/aom_dsp/sad_av1.c vendored
View file

@ -75,10 +75,10 @@ MASKSADMxN(8, 32)
MASKSADMxN(32, 8)
MASKSADMxN(16, 64)
MASKSADMxN(64, 16)
/* clang-format on */
/* clang-format on */
static INLINE
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE
unsigned int highbd_masked_sad(const uint8_t *src8, int src_stride,
const uint8_t *a8, int a_stride,
const uint8_t *b8, int b_stride,
@ -141,6 +141,7 @@ HIGHBD_MASKSADMXN(8, 32)
HIGHBD_MASKSADMXN(32, 8)
HIGHBD_MASKSADMXN(16, 64)
HIGHBD_MASKSADMXN(64, 16)
#endif // CONFIG_AV1_HIGHBITDEPTH
// pre: predictor being evaluated
// wsrc: target weighted prediction (has been *4096 to keep precision)
@ -193,9 +194,10 @@ OBMCSADMxN(8, 32)
OBMCSADMxN(32, 8)
OBMCSADMxN(16, 64)
OBMCSADMxN(64, 16)
/* clang-format on */
/* clang-format on */
static INLINE
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE
unsigned int highbd_obmc_sad(const uint8_t *pre8, int pre_stride,
const int32_t *wsrc, const int32_t *mask,
int width, int height) {
@ -246,3 +248,4 @@ HIGHBD_OBMCSADMXN(32, 8)
HIGHBD_OBMCSADMXN(16, 64)
HIGHBD_OBMCSADMXN(64, 16)
/* clang-format on */
#endif // CONFIG_AV1_HIGHBITDEPTH

5
third_party/aom/aom_dsp/simd/v128_intrinsics_x86.h vendored
View file

@ -89,7 +89,8 @@ SIMD_INLINE v128 v128_dup_32(uint32_t x) { return _mm_set1_epi32(x); }
SIMD_INLINE v128 v128_dup_64(uint64_t x) {
// _mm_set_pi64x and _mm_cvtsi64x_si64 missing in some compilers
return _mm_set_epi32(x >> 32, (uint32_t)x, x >> 32, (uint32_t)x);
return _mm_set_epi32((uint32_t)(x >> 32), (uint32_t)x, (uint32_t)(x >> 32),
(uint32_t)x);
}
SIMD_INLINE v128 v128_add_8(v128 a, v128 b) { return _mm_add_epi8(a, b); }
@ -538,7 +539,7 @@ SIMD_INLINE v128 v128_shl_8(v128 a, unsigned int c) {
}
SIMD_INLINE v128 v128_shr_u8(v128 a, unsigned int c) {
return _mm_and_si128(_mm_set1_epi8(0xff >> c),
return _mm_and_si128(_mm_set1_epi8((char)(0xff >> c)),
_mm_srl_epi16(a, _mm_cvtsi32_si128(c)));
}

2
third_party/aom/aom_dsp/simd/v256_intrinsics_x86.h vendored
View file

@ -603,7 +603,7 @@ SIMD_INLINE v256 v256_shl_8(v256 a, unsigned int c) {
}
SIMD_INLINE v256 v256_shr_u8(v256 a, unsigned int c) {
return _mm256_and_si256(_mm256_set1_epi8(0xff >> c),
return _mm256_and_si256(_mm256_set1_epi8((char)(0xff >> c)),
_mm256_srl_epi16(a, _mm_cvtsi32_si128(c)));
}

33
third_party/aom/aom_dsp/simd/v64_intrinsics_c.h vendored
View file

@ -171,14 +171,14 @@ SIMD_INLINE c_v64 c_v64_dup_32(uint32_t x) {
SIMD_INLINE c_v64 c_v64_add_8(c_v64 a, c_v64 b) {
c_v64 t;
int c;
for (c = 0; c < 8; c++) t.u8[c] = a.u8[c] + b.u8[c];
for (c = 0; c < 8; c++) t.u8[c] = (uint8_t)(a.u8[c] + b.u8[c]);
return t;
}
SIMD_INLINE c_v64 c_v64_add_16(c_v64 a, c_v64 b) {
c_v64 t;
int c;
for (c = 0; c < 4; c++) t.u16[c] = a.u16[c] + b.u16[c];
for (c = 0; c < 4; c++) t.u16[c] = (uint16_t)(a.u16[c] + b.u16[c]);
return t;
}
@ -228,7 +228,7 @@ SIMD_INLINE c_v64 c_v64_add_32(c_v64 a, c_v64 b) {
SIMD_INLINE c_v64 c_v64_sub_8(c_v64 a, c_v64 b) {
c_v64 t;
int c;
for (c = 0; c < 8; c++) t.u8[c] = a.u8[c] - b.u8[c];
for (c = 0; c < 8; c++) t.u8[c] = (uint8_t)(a.u8[c] - b.u8[c]);
return t;
}
@ -252,7 +252,7 @@ SIMD_INLINE c_v64 c_v64_ssub_s8(c_v64 a, c_v64 b) {
SIMD_INLINE c_v64 c_v64_sub_16(c_v64 a, c_v64 b) {
c_v64 t;
int c;
for (c = 0; c < 4; c++) t.u16[c] = a.u16[c] - b.u16[c];
for (c = 0; c < 4; c++) t.u16[c] = (uint16_t)(a.u16[c] - b.u16[c]);
return t;
}
@ -288,14 +288,15 @@ SIMD_INLINE c_v64 c_v64_abs_s16(c_v64 a) {
c_v64 t;
int c;
for (c = 0; c < 4; c++)
t.u16[c] = (int16_t)a.u16[c] > 0 ? a.u16[c] : -a.u16[c];
t.u16[c] = (uint16_t)((int16_t)a.u16[c] > 0 ? a.u16[c] : -a.u16[c]);
return t;
}
SIMD_INLINE c_v64 c_v64_abs_s8(c_v64 a) {
c_v64 t;
int c;
for (c = 0; c < 8; c++) t.u8[c] = (int8_t)a.u8[c] > 0 ? a.u8[c] : -a.u8[c];
for (c = 0; c < 8; c++)
t.u8[c] = (uint8_t)((int8_t)a.u8[c] > 0 ? a.u8[c] : -a.u8[c]);
return t;
}
@ -526,14 +527,14 @@ SIMD_INLINE c_v64 c_v64_pack_s16_s8(c_v64 a, c_v64 b) {
a = b;
b = u;
}
t.u8[7] = a.s16[3] > 127 ? 127 : a.s16[3] < -128 ? 128 : a.s16[3];
t.u8[6] = a.s16[2] > 127 ? 127 : a.s16[2] < -128 ? 128 : a.s16[2];
t.u8[5] = a.s16[1] > 127 ? 127 : a.s16[1] < -128 ? 128 : a.s16[1];
t.u8[4] = a.s16[0] > 127 ? 127 : a.s16[0] < -128 ? 128 : a.s16[0];
t.u8[3] = b.s16[3] > 127 ? 127 : b.s16[3] < -128 ? 128 : b.s16[3];
t.u8[2] = b.s16[2] > 127 ? 127 : b.s16[2] < -128 ? 128 : b.s16[2];
t.u8[1] = b.s16[1] > 127 ? 127 : b.s16[1] < -128 ? 128 : b.s16[1];
t.u8[0] = b.s16[0] > 127 ? 127 : b.s16[0] < -128 ? 128 : b.s16[0];
t.u8[7] = (uint8_t)(a.s16[3] > 127 ? 127 : a.s16[3] < -128 ? 128 : a.s16[3]);
t.u8[6] = (uint8_t)(a.s16[2] > 127 ? 127 : a.s16[2] < -128 ? 128 : a.s16[2]);
t.u8[5] = (uint8_t)(a.s16[1] > 127 ? 127 : a.s16[1] < -128 ? 128 : a.s16[1]);
t.u8[4] = (uint8_t)(a.s16[0] > 127 ? 127 : a.s16[0] < -128 ? 128 : a.s16[0]);
t.u8[3] = (uint8_t)(b.s16[3] > 127 ? 127 : b.s16[3] < -128 ? 128 : b.s16[3]);
t.u8[2] = (uint8_t)(b.s16[2] > 127 ? 127 : b.s16[2] < -128 ? 128 : b.s16[2]);
t.u8[1] = (uint8_t)(b.s16[1] > 127 ? 127 : b.s16[1] < -128 ? 128 : b.s16[1]);
t.u8[0] = (uint8_t)(b.s16[0] > 127 ? 127 : b.s16[0] < -128 ? 128 : b.s16[0]);
return t;
}
@ -817,7 +818,7 @@ SIMD_INLINE c_v64 c_v64_shl_8(c_v64 a, unsigned int n) {
fprintf(stderr, "Error: Undefined u8 shift left %d\n", n);
abort();
}
for (c = 0; c < 8; c++) t.s8[c] = a.u8[c] << n;
for (c = 0; c < 8; c++) t.s8[c] = (int8_t)(a.u8[c] << n);
return t;
}
@ -850,7 +851,7 @@ SIMD_INLINE c_v64 c_v64_shl_16(c_v64 a, unsigned int n) {
fprintf(stderr, "Error: Undefined u16 shift left %d\n", n);
abort();
}
for (c = 0; c < 4; c++) t.u16[c] = a.u16[c] << n;
for (c = 0; c < 4; c++) t.u16[c] = (uint16_t)(a.u16[c] << n);
return t;
}

4
third_party/aom/aom_dsp/simd/v64_intrinsics_x86.h vendored
View file

@ -433,12 +433,12 @@ SIMD_INLINE v64 v64_cmplt_s16(v64 a, v64 b) { return _mm_cmplt_epi16(a, b); }
SIMD_INLINE v64 v64_cmpeq_16(v64 a, v64 b) { return _mm_cmpeq_epi16(a, b); }
SIMD_INLINE v64 v64_shl_8(v64 a, unsigned int c) {
return _mm_and_si128(_mm_set1_epi8((uint8_t)(0xff << c)),
return _mm_and_si128(_mm_set1_epi8((char)(0xff << c)),
_mm_sll_epi16(a, _mm_cvtsi32_si128(c)));
}
SIMD_INLINE v64 v64_shr_u8(v64 a, unsigned int c) {
return _mm_and_si128(_mm_set1_epi8(0xff >> c),
return _mm_and_si128(_mm_set1_epi8((char)(0xff >> c)),
_mm_srl_epi16(a, _mm_cvtsi32_si128(c)));
}

2
third_party/aom/aom_dsp/sse.c vendored
View file

@ -33,6 +33,7 @@ int64_t aom_sse_c(const uint8_t *a, int a_stride, const uint8_t *b,
return sse;
}
#if CONFIG_AV1_HIGHBITDEPTH
int64_t aom_highbd_sse_c(const uint8_t *a8, int a_stride, const uint8_t *b8,
int b_stride, int width, int height) {
int y, x;
@ -50,3 +51,4 @@ int64_t aom_highbd_sse_c(const uint8_t *a8, int a_stride, const uint8_t *b8,
}
return sse;
}
#endif

2
third_party/aom/aom_dsp/ssim.c vendored
View file

@ -49,6 +49,7 @@ void aom_ssim_parms_8x8_c(const uint8_t *s, int sp, const uint8_t *r, int rp,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_ssim_parms_8x8_c(const uint16_t *s, int sp, const uint16_t *r,
int rp, uint32_t *sum_s, uint32_t *sum_r,
uint32_t *sum_sq_s, uint32_t *sum_sq_r,
@ -64,6 +65,7 @@ void aom_highbd_ssim_parms_8x8_c(const uint16_t *s, int sp, const uint16_t *r,
}
}
}
#endif
static const int64_t cc1 = 26634; // (64^2*(.01*255)^2
static const int64_t cc2 = 239708; // (64^2*(.03*255)^2

2
third_party/aom/aom_dsp/subtract.c vendored
View file

@ -32,6 +32,7 @@ void aom_subtract_block_c(int rows, int cols, int16_t *diff,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_subtract_block_c(int rows, int cols, int16_t *diff,
ptrdiff_t diff_stride, const uint8_t *src8,
ptrdiff_t src_stride, const uint8_t *pred8,
@ -51,3 +52,4 @@ void aom_highbd_subtract_block_c(int rows, int cols, int16_t *diff,
src += src_stride;
}
}
#endif

56
third_party/aom/aom_dsp/variance.c vendored
View file

@ -339,30 +339,25 @@ void aom_upsampled_pred_c(MACROBLOCKD *xd, const AV1_COMMON *const cm,
pre_buf->buf0 + (pos_y >> SCALE_SUBPEL_BITS) * pre_buf->stride +
(pos_x >> SCALE_SUBPEL_BITS);
InterPredParams inter_pred_params;
const SubpelParams subpel_params = { sf->x_step_q4, sf->y_step_q4,
pos_x & SCALE_SUBPEL_MASK,
pos_y & SCALE_SUBPEL_MASK };
// Get warp types.
const WarpedMotionParams *const wm =
&xd->global_motion[mi->ref_frame[ref_num]];
const int is_global = is_global_mv_block(mi, wm->wmtype);
WarpTypesAllowed warp_types;
warp_types.global_warp_allowed = is_global;
warp_types.local_warp_allowed = mi->motion_mode == WARPED_CAUSAL;
// Get convolve parameters.
ConvolveParams conv_params = get_conv_params(0, plane, xd->bd);
const InterpFilters filters =
inter_pred_params.conv_params = get_conv_params(0, plane, xd->bd);
const int_interpfilters filters =
av1_broadcast_interp_filter(EIGHTTAP_REGULAR);
av1_init_inter_params(
&inter_pred_params, width, height, mi_y >> pd->subsampling_y,
mi_x >> pd->subsampling_x, pd->subsampling_x, pd->subsampling_y,
xd->bd, is_cur_buf_hbd(xd), mi->use_intrabc, sf, filters);
// Get the inter predictor.
const int build_for_obmc = 0;
av1_make_inter_predictor(pre, pre_buf->stride, comp_pred, width,
&subpel_params, sf, width, height, &conv_params,
filters, &warp_types, mi_x >> pd->subsampling_x,
mi_y >> pd->subsampling_y, plane, ref_num, mi,
build_for_obmc, xd, cm->allow_warped_motion);
&inter_pred_params, &subpel_params);
return;
}
@ -468,6 +463,7 @@ void aom_dist_wtd_comp_avg_upsampled_pred_c(
}
}
#if CONFIG_AV1_HIGHBITDEPTH
static void highbd_variance64(const uint8_t *a8, int a_stride,
const uint8_t *b8, int b_stride, int w, int h,
uint64_t *sse, int64_t *sum) {
@ -931,30 +927,25 @@ void aom_highbd_upsampled_pred_c(MACROBLOCKD *xd,
pre_buf->buf0 + (pos_y >> SCALE_SUBPEL_BITS) * pre_buf->stride +
(pos_x >> SCALE_SUBPEL_BITS);
InterPredParams inter_pred_params;
const SubpelParams subpel_params = { sf->x_step_q4, sf->y_step_q4,
pos_x & SCALE_SUBPEL_MASK,
pos_y & SCALE_SUBPEL_MASK };
// Get warp types.
const WarpedMotionParams *const wm =
&xd->global_motion[mi->ref_frame[ref_num]];
const int is_global = is_global_mv_block(mi, wm->wmtype);
WarpTypesAllowed warp_types;
warp_types.global_warp_allowed = is_global;
warp_types.local_warp_allowed = mi->motion_mode == WARPED_CAUSAL;
// Get convolve parameters.
ConvolveParams conv_params = get_conv_params(0, plane, xd->bd);
const InterpFilters filters =
inter_pred_params.conv_params = get_conv_params(0, plane, xd->bd);
const int_interpfilters filters =
av1_broadcast_interp_filter(EIGHTTAP_REGULAR);
av1_init_inter_params(
&inter_pred_params, width, height, mi_y >> pd->subsampling_y,
mi_x >> pd->subsampling_x, pd->subsampling_x, pd->subsampling_y,
xd->bd, is_cur_buf_hbd(xd), mi->use_intrabc, sf, filters);
// Get the inter predictor.
const int build_for_obmc = 0;
av1_make_inter_predictor(pre, pre_buf->stride, comp_pred8, width,
&subpel_params, sf, width, height, &conv_params,
filters, &warp_types, mi_x >> pd->subsampling_x,
mi_y >> pd->subsampling_y, plane, ref_num, mi,
build_for_obmc, xd, cm->allow_warped_motion);
&inter_pred_params, &subpel_params);
return;
}
@ -1070,6 +1061,7 @@ void aom_highbd_dist_wtd_comp_avg_upsampled_pred_c(
pred += width;
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH
void aom_comp_mask_pred_c(uint8_t *comp_pred, const uint8_t *pred, int width,
int height, const uint8_t *ref, int ref_stride,
@ -1153,6 +1145,7 @@ MASK_SUBPIX_VAR(32, 8)
MASK_SUBPIX_VAR(16, 64)
MASK_SUBPIX_VAR(64, 16)
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_comp_mask_pred_c(uint8_t *comp_pred8, const uint8_t *pred8,
int width, int height, const uint8_t *ref8,
int ref_stride, const uint8_t *mask,
@ -1277,6 +1270,7 @@ HIGHBD_MASK_SUBPIX_VAR(8, 32)
HIGHBD_MASK_SUBPIX_VAR(32, 8)
HIGHBD_MASK_SUBPIX_VAR(16, 64)
HIGHBD_MASK_SUBPIX_VAR(64, 16)
#endif // CONFIG_AV1_HIGHBITDEPTH
static INLINE void obmc_variance(const uint8_t *pre, int pre_stride,
const int32_t *wsrc, const int32_t *mask,
@ -1384,6 +1378,7 @@ OBMC_SUBPIX_VAR(16, 64)
OBMC_VAR(64, 16)
OBMC_SUBPIX_VAR(64, 16)
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE void highbd_obmc_variance64(const uint8_t *pre8, int pre_stride,
const int32_t *wsrc,
const int32_t *mask, int w, int h,
@ -1575,3 +1570,4 @@ HIGHBD_OBMC_VAR(16, 64)
HIGHBD_OBMC_SUBPIX_VAR(16, 64)
HIGHBD_OBMC_VAR(64, 16)
HIGHBD_OBMC_SUBPIX_VAR(64, 16)
#endif // CONFIG_AV1_HIGHBITDEPTH

3
third_party/aom/aom_dsp/x86/aom_asm_stubs.c vendored
View file

@ -49,6 +49,7 @@ filter8_1dfunction aom_filter_block1d4_h2_sse2;
FUN_CONV_1D(horiz, x_step_q4, filter_x, h, src, , sse2);
FUN_CONV_1D(vert, y_step_q4, filter_y, v, src - src_stride * 3, , sse2);
#if CONFIG_AV1_HIGHBITDEPTH
highbd_filter8_1dfunction aom_highbd_filter_block1d16_v8_sse2;
highbd_filter8_1dfunction aom_highbd_filter_block1d16_h8_sse2;
highbd_filter8_1dfunction aom_highbd_filter_block1d8_v8_sse2;
@ -90,5 +91,5 @@ highbd_filter8_1dfunction aom_highbd_filter_block1d4_h2_sse2;
// int w, int h, int bd);
HIGH_FUN_CONV_1D(horiz, x_step_q4, filter_x, h, src, , sse2);
HIGH_FUN_CONV_1D(vert, y_step_q4, filter_y, v, src - src_stride * 3, , sse2);
#endif
#endif // HAVE_SSE2

214
third_party/aom/aom_dsp/x86/avg_intrin_avx2.c vendored
View file

@ -224,17 +224,217 @@ void aom_hadamard_32x32_avx2(const int16_t *src_diff, ptrdiff_t src_stride,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
static void highbd_hadamard_col8_avx2(__m256i *in, int iter) {
__m256i a0 = in[0];
__m256i a1 = in[1];
__m256i a2 = in[2];
__m256i a3 = in[3];
__m256i a4 = in[4];
__m256i a5 = in[5];
__m256i a6 = in[6];
__m256i a7 = in[7];
__m256i b0 = _mm256_add_epi32(a0, a1);
__m256i b1 = _mm256_sub_epi32(a0, a1);
__m256i b2 = _mm256_add_epi32(a2, a3);
__m256i b3 = _mm256_sub_epi32(a2, a3);
__m256i b4 = _mm256_add_epi32(a4, a5);
__m256i b5 = _mm256_sub_epi32(a4, a5);
__m256i b6 = _mm256_add_epi32(a6, a7);
__m256i b7 = _mm256_sub_epi32(a6, a7);
a0 = _mm256_add_epi32(b0, b2);
a1 = _mm256_add_epi32(b1, b3);
a2 = _mm256_sub_epi32(b0, b2);
a3 = _mm256_sub_epi32(b1, b3);
a4 = _mm256_add_epi32(b4, b6);
a5 = _mm256_add_epi32(b5, b7);
a6 = _mm256_sub_epi32(b4, b6);
a7 = _mm256_sub_epi32(b5, b7);
if (iter == 0) {
b0 = _mm256_add_epi32(a0, a4);
b7 = _mm256_add_epi32(a1, a5);
b3 = _mm256_add_epi32(a2, a6);
b4 = _mm256_add_epi32(a3, a7);
b2 = _mm256_sub_epi32(a0, a4);
b6 = _mm256_sub_epi32(a1, a5);
b1 = _mm256_sub_epi32(a2, a6);
b5 = _mm256_sub_epi32(a3, a7);
a0 = _mm256_unpacklo_epi32(b0, b1);
a1 = _mm256_unpacklo_epi32(b2, b3);
a2 = _mm256_unpackhi_epi32(b0, b1);
a3 = _mm256_unpackhi_epi32(b2, b3);
a4 = _mm256_unpacklo_epi32(b4, b5);
a5 = _mm256_unpacklo_epi32(b6, b7);
a6 = _mm256_unpackhi_epi32(b4, b5);
a7 = _mm256_unpackhi_epi32(b6, b7);
b0 = _mm256_unpacklo_epi64(a0, a1);
b1 = _mm256_unpacklo_epi64(a4, a5);
b2 = _mm256_unpackhi_epi64(a0, a1);
b3 = _mm256_unpackhi_epi64(a4, a5);
b4 = _mm256_unpacklo_epi64(a2, a3);
b5 = _mm256_unpacklo_epi64(a6, a7);
b6 = _mm256_unpackhi_epi64(a2, a3);
b7 = _mm256_unpackhi_epi64(a6, a7);
in[0] = _mm256_permute2x128_si256(b0, b1, 0x20);
in[1] = _mm256_permute2x128_si256(b0, b1, 0x31);
in[2] = _mm256_permute2x128_si256(b2, b3, 0x20);
in[3] = _mm256_permute2x128_si256(b2, b3, 0x31);
in[4] = _mm256_permute2x128_si256(b4, b5, 0x20);
in[5] = _mm256_permute2x128_si256(b4, b5, 0x31);
in[6] = _mm256_permute2x128_si256(b6, b7, 0x20);
in[7] = _mm256_permute2x128_si256(b6, b7, 0x31);
} else {
in[0] = _mm256_add_epi32(a0, a4);
in[7] = _mm256_add_epi32(a1, a5);
in[3] = _mm256_add_epi32(a2, a6);
in[4] = _mm256_add_epi32(a3, a7);
in[2] = _mm256_sub_epi32(a0, a4);
in[6] = _mm256_sub_epi32(a1, a5);
in[1] = _mm256_sub_epi32(a2, a6);
in[5] = _mm256_sub_epi32(a3, a7);
}
}
void aom_highbd_hadamard_8x8_avx2(const int16_t *src_diff, ptrdiff_t src_stride,
tran_low_t *coeff) {
__m128i src16[8];
__m256i src32[8];
src16[0] = _mm_loadu_si128((const __m128i *)src_diff);
src16[1] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src16[2] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src16[3] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src16[4] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src16[5] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src16[6] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src16[7] = _mm_loadu_si128((const __m128i *)(src_diff += src_stride));
src32[0] = _mm256_cvtepi16_epi32(src16[0]);
src32[1] = _mm256_cvtepi16_epi32(src16[1]);
src32[2] = _mm256_cvtepi16_epi32(src16[2]);
src32[3] = _mm256_cvtepi16_epi32(src16[3]);
src32[4] = _mm256_cvtepi16_epi32(src16[4]);
src32[5] = _mm256_cvtepi16_epi32(src16[5]);
src32[6] = _mm256_cvtepi16_epi32(src16[6]);
src32[7] = _mm256_cvtepi16_epi32(src16[7]);
highbd_hadamard_col8_avx2(src32, 0);
highbd_hadamard_col8_avx2(src32, 1);
_mm256_storeu_si256((__m256i *)coeff, src32[0]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[1]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[2]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[3]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[4]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[5]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[6]);
coeff += 8;
_mm256_storeu_si256((__m256i *)coeff, src32[7]);
}
void aom_highbd_hadamard_16x16_avx2(const int16_t *src_diff,
ptrdiff_t src_stride, tran_low_t *coeff) {
int idx;
tran_low_t *t_coeff = coeff;
for (idx = 0; idx < 4; ++idx) {
const int16_t *src_ptr =
src_diff + (idx >> 1) * 8 * src_stride + (idx & 0x01) * 8;
aom_highbd_hadamard_8x8_avx2(src_ptr, src_stride, t_coeff + idx * 64);
}
for (idx = 0; idx < 64; idx += 8) {
__m256i coeff0 = _mm256_loadu_si256((const __m256i *)t_coeff);
__m256i coeff1 = _mm256_loadu_si256((const __m256i *)(t_coeff + 64));
__m256i coeff2 = _mm256_loadu_si256((const __m256i *)(t_coeff + 128));
__m256i coeff3 = _mm256_loadu_si256((const __m256i *)(t_coeff + 192));
__m256i b0 = _mm256_add_epi32(coeff0, coeff1);
__m256i b1 = _mm256_sub_epi32(coeff0, coeff1);
__m256i b2 = _mm256_add_epi32(coeff2, coeff3);
__m256i b3 = _mm256_sub_epi32(coeff2, coeff3);
b0 = _mm256_srai_epi32(b0, 1);
b1 = _mm256_srai_epi32(b1, 1);
b2 = _mm256_srai_epi32(b2, 1);
b3 = _mm256_srai_epi32(b3, 1);
coeff0 = _mm256_add_epi32(b0, b2);
coeff1 = _mm256_add_epi32(b1, b3);
coeff2 = _mm256_sub_epi32(b0, b2);
coeff3 = _mm256_sub_epi32(b1, b3);
_mm256_storeu_si256((__m256i *)coeff, coeff0);
_mm256_storeu_si256((__m256i *)(coeff + 64), coeff1);
_mm256_storeu_si256((__m256i *)(coeff + 128), coeff2);
_mm256_storeu_si256((__m256i *)(coeff + 192), coeff3);
coeff += 8;
t_coeff += 8;
}
}
void aom_highbd_hadamard_32x32_avx2(const int16_t *src_diff,
ptrdiff_t src_stride, tran_low_t *coeff) {
int idx;
tran_low_t *t_coeff = coeff;
for (idx = 0; idx < 4; ++idx) {
const int16_t *src_ptr =
src_diff + (idx >> 1) * 16 * src_stride + (idx & 0x01) * 16;
aom_highbd_hadamard_16x16_avx2(src_ptr, src_stride, t_coeff + idx * 256);
}
for (idx = 0; idx < 256; idx += 8) {
__m256i coeff0 = _mm256_loadu_si256((const __m256i *)t_coeff);
__m256i coeff1 = _mm256_loadu_si256((const __m256i *)(t_coeff + 256));
__m256i coeff2 = _mm256_loadu_si256((const __m256i *)(t_coeff + 512));
__m256i coeff3 = _mm256_loadu_si256((const __m256i *)(t_coeff + 768));
__m256i b0 = _mm256_add_epi32(coeff0, coeff1);
__m256i b1 = _mm256_sub_epi32(coeff0, coeff1);
__m256i b2 = _mm256_add_epi32(coeff2, coeff3);
__m256i b3 = _mm256_sub_epi32(coeff2, coeff3);
b0 = _mm256_srai_epi32(b0, 2);
b1 = _mm256_srai_epi32(b1, 2);
b2 = _mm256_srai_epi32(b2, 2);
b3 = _mm256_srai_epi32(b3, 2);
coeff0 = _mm256_add_epi32(b0, b2);
coeff1 = _mm256_add_epi32(b1, b3);
coeff2 = _mm256_sub_epi32(b0, b2);
coeff3 = _mm256_sub_epi32(b1, b3);
_mm256_storeu_si256((__m256i *)coeff, coeff0);
_mm256_storeu_si256((__m256i *)(coeff + 256), coeff1);
_mm256_storeu_si256((__m256i *)(coeff + 512), coeff2);
_mm256_storeu_si256((__m256i *)(coeff + 768), coeff3);
coeff += 8;
t_coeff += 8;
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH
int aom_satd_avx2(const tran_low_t *coeff, int length) {
const __m256i one = _mm256_set1_epi16(1);
__m256i accum = _mm256_setzero_si256();
int i;
for (i = 0; i < length; i += 16) {
const __m256i src_line = load_tran_low(coeff);
const __m256i abs = _mm256_abs_epi16(src_line);
const __m256i sum = _mm256_madd_epi16(abs, one);
accum = _mm256_add_epi32(accum, sum);
coeff += 16;
for (i = 0; i < length; i += 8, coeff += 8) {
const __m256i src_line = _mm256_loadu_si256((const __m256i *)coeff);
const __m256i abs = _mm256_abs_epi32(src_line);
accum = _mm256_add_epi32(accum, abs);
}
{ // 32 bit horizontal add

72
third_party/aom/aom_dsp/x86/avg_intrin_sse2.c vendored
View file

@ -406,3 +406,75 @@ int aom_satd_sse2(const tran_low_t *coeff, int length) {
return _mm_cvtsi128_si32(accum);
}
void aom_int_pro_row_sse2(int16_t *hbuf, const uint8_t *ref,
const int ref_stride, const int height) {
int idx = 1;
__m128i zero = _mm_setzero_si128();
__m128i src_line = _mm_loadu_si128((const __m128i *)ref);
__m128i s0 = _mm_unpacklo_epi8(src_line, zero);
__m128i s1 = _mm_unpackhi_epi8(src_line, zero);
__m128i t0, t1;
int height_1 = height - 1;
ref += ref_stride;
do {
src_line = _mm_loadu_si128((const __m128i *)ref);
t0 = _mm_unpacklo_epi8(src_line, zero);
t1 = _mm_unpackhi_epi8(src_line, zero);
s0 = _mm_adds_epu16(s0, t0);
s1 = _mm_adds_epu16(s1, t1);
ref += ref_stride;
src_line = _mm_loadu_si128((const __m128i *)ref);
t0 = _mm_unpacklo_epi8(src_line, zero);
t1 = _mm_unpackhi_epi8(src_line, zero);
s0 = _mm_adds_epu16(s0, t0);
s1 = _mm_adds_epu16(s1, t1);
ref += ref_stride;
idx += 2;
} while (idx < height_1);
src_line = _mm_loadu_si128((const __m128i *)ref);
t0 = _mm_unpacklo_epi8(src_line, zero);
t1 = _mm_unpackhi_epi8(src_line, zero);
s0 = _mm_adds_epu16(s0, t0);
s1 = _mm_adds_epu16(s1, t1);
if (height == 128) {
s0 = _mm_srai_epi16(s0, 6);
s1 = _mm_srai_epi16(s1, 6);
} else if (height == 64) {
s0 = _mm_srai_epi16(s0, 5);
s1 = _mm_srai_epi16(s1, 5);
} else if (height == 32) {
s0 = _mm_srai_epi16(s0, 4);
s1 = _mm_srai_epi16(s1, 4);
} else {
assert(height == 16);
s0 = _mm_srai_epi16(s0, 3);
s1 = _mm_srai_epi16(s1, 3);
}
_mm_storeu_si128((__m128i *)hbuf, s0);
hbuf += 8;
_mm_storeu_si128((__m128i *)hbuf, s1);
}
int16_t aom_int_pro_col_sse2(const uint8_t *ref, const int width) {
__m128i zero = _mm_setzero_si128();
__m128i src_line = _mm_loadu_si128((const __m128i *)ref);
__m128i s0 = _mm_sad_epu8(src_line, zero);
__m128i s1;
int i;
for (i = 16; i < width; i += 16) {
ref += 16;
src_line = _mm_loadu_si128((const __m128i *)ref);
s1 = _mm_sad_epu8(src_line, zero);
s0 = _mm_adds_epu16(s0, s1);
}
s1 = _mm_srli_si128(s0, 8);
s0 = _mm_adds_epu16(s0, s1);
return _mm_extract_epi16(s0, 0);
}

2
third_party/aom/aom_dsp/x86/blend_a64_hmask_sse4.c vendored
View file

@ -24,6 +24,7 @@ void aom_blend_a64_hmask_sse4_1(uint8_t *dst, uint32_t dst_stride,
src1_stride, mask, 0, w, h, 0, 0);
}
#if CONFIG_AV1_HIGHBITDEPTH
void aom_highbd_blend_a64_hmask_sse4_1(
uint8_t *dst_8, uint32_t dst_stride, const uint8_t *src0_8,
uint32_t src0_stride, const uint8_t *src1_8, uint32_t src1_stride,
@ -32,3 +33,4 @@ void aom_highbd_blend_a64_hmask_sse4_1(
src1_8, src1_stride, mask, 0, w, h, 0, 0,
bd);
}
#endif

12
third_party/aom/aom_dsp/x86/blend_a64_mask_avx2.c vendored
View file

@ -870,7 +870,7 @@ void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride,
const uint8_t *src0, uint32_t src0_stride,
const uint8_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int w,
int h, int subx, int suby) {
int h, int subw, int subh) {
assert(IMPLIES(src0 == dst, src0_stride == dst_stride));
assert(IMPLIES(src1 == dst, src1_stride == dst_stride));
@ -881,15 +881,15 @@ void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride,
if (UNLIKELY((h | w) & 3)) { // if (w <= 2 || h <= 2)
aom_blend_a64_mask_c(dst, dst_stride, src0, src0_stride, src1, src1_stride,
mask, mask_stride, w, h, subx, suby);
mask, mask_stride, w, h, subw, subh);
} else {
if (subx & suby) {
if (subw & subh) {
blend_a64_mask_sx_sy_avx2(dst, dst_stride, src0, src0_stride, src1,
src1_stride, mask, mask_stride, w, h);
} else if (subx) {
} else if (subw) {
blend_a64_mask_sx_avx2(dst, dst_stride, src0, src0_stride, src1,
src1_stride, mask, mask_stride, w, h);
} else if (suby) {
} else if (subh) {
blend_a64_mask_sy_avx2(dst, dst_stride, src0, src0_stride, src1,
src1_stride, mask, mask_stride, w, h);
} else {
@ -899,6 +899,7 @@ void aom_blend_a64_mask_avx2(uint8_t *dst, uint32_t dst_stride,
}
}
#if CONFIG_AV1_HIGHBITDEPTH
//////////////////////////////////////////////////////////////////////////////
// aom_highbd_blend_a64_d16_mask_avx2()
//////////////////////////////////////////////////////////////////////////////
@ -1370,3 +1371,4 @@ void aom_highbd_blend_a64_d16_mask_avx2(
subh, conv_params, bd);
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH

42
third_party/aom/aom_dsp/x86/blend_a64_mask_sse4.c vendored
View file

@ -339,8 +339,8 @@ static void blend_a64_mask_sx_sy_w16n_sse4_1(
uint8_t *dst, uint32_t dst_stride, const uint8_t *src0,
uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int w, int h) {
const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff, 0,
0xff, 0, 0xff, 0, 0xff, 0, 0xff);
const __m128i v_zmask_b =
_mm_set_epi8(0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1);
const __m128i v_maxval_b = _mm_set1_epi8(AOM_BLEND_A64_MAX_ALPHA);
const __m128i _r = _mm_set1_epi16(1 << (15 - AOM_BLEND_A64_ROUND_BITS));
do {
@ -386,7 +386,7 @@ void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride,
const uint8_t *src0, uint32_t src0_stride,
const uint8_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int w,
int h, int subx, int suby) {
int h, int subw, int subh) {
typedef void (*blend_fn)(
uint8_t * dst, uint32_t dst_stride, const uint8_t *src0,
uint32_t src0_stride, const uint8_t *src1, uint32_t src1_stride,
@ -415,14 +415,15 @@ void aom_blend_a64_mask_sse4_1(uint8_t *dst, uint32_t dst_stride,
if (UNLIKELY((h | w) & 3)) { // if (w <= 2 || h <= 2)
aom_blend_a64_mask_c(dst, dst_stride, src0, src0_stride, src1, src1_stride,
mask, mask_stride, w, h, subx, suby);
mask, mask_stride, w, h, subw, subh);
} else {
blend[(w >> 2) & 3][subx != 0][suby != 0](dst, dst_stride, src0,
blend[(w >> 2) & 3][subw != 0][subh != 0](dst, dst_stride, src0,
src0_stride, src1, src1_stride,
mask, mask_stride, w, h);
}
}
#if CONFIG_AV1_HIGHBITDEPTH
//////////////////////////////////////////////////////////////////////////////
// No sub-sampling
//////////////////////////////////////////////////////////////////////////////
@ -518,8 +519,8 @@ static INLINE void blend_a64_mask_bn_sx_w4_sse4_1(
uint16_t *dst, uint32_t dst_stride, const uint16_t *src0,
uint32_t src0_stride, const uint16_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int h, blend_unit_fn blend) {
const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff, 0,
0xff, 0, 0xff, 0, 0xff, 0, 0xff);
const __m128i v_zmask_b =
_mm_set_epi8(0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1);
const __m128i v_maxval_w = _mm_set1_epi16(AOM_BLEND_A64_MAX_ALPHA);
do {
@ -565,8 +566,8 @@ static INLINE void blend_a64_mask_bn_sx_w8n_sse4_1(
uint32_t src0_stride, const uint16_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int w, int h,
blend_unit_fn blend) {
const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff, 0,
0xff, 0, 0xff, 0, 0xff, 0, 0xff);
const __m128i v_zmask_b =
_mm_set_epi8(0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1);
const __m128i v_maxval_w = _mm_set1_epi16(AOM_BLEND_A64_MAX_ALPHA);
do {
@ -710,8 +711,8 @@ static INLINE void blend_a64_mask_bn_sx_sy_w4_sse4_1(
uint16_t *dst, uint32_t dst_stride, const uint16_t *src0,
uint32_t src0_stride, const uint16_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int h, blend_unit_fn blend) {
const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff, 0,
0xff, 0, 0xff, 0, 0xff, 0, 0xff);
const __m128i v_zmask_b =
_mm_set_epi8(0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1);
const __m128i v_maxval_w = _mm_set1_epi16(AOM_BLEND_A64_MAX_ALPHA);
do {
@ -762,8 +763,8 @@ static INLINE void blend_a64_mask_bn_sx_sy_w8n_sse4_1(
uint32_t src0_stride, const uint16_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int w, int h,
blend_unit_fn blend) {
const __m128i v_zmask_b = _mm_set_epi8(0, 0xff, 0, 0xff, 0, 0xff, 0, 0xff, 0,
0xff, 0, 0xff, 0, 0xff, 0, 0xff);
const __m128i v_zmask_b =
_mm_set_epi8(0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1);
const __m128i v_maxval_w = _mm_set1_epi16(AOM_BLEND_A64_MAX_ALPHA);
do {
@ -812,20 +813,19 @@ static void blend_a64_mask_b12_sx_sy_w8n_sse4_1(
//////////////////////////////////////////////////////////////////////////////
// Dispatch
//////////////////////////////////////////////////////////////////////////////
void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst_8, uint32_t dst_stride,
const uint8_t *src0_8,
uint32_t src0_stride,
const uint8_t *src1_8,
uint32_t src1_stride, const uint8_t *mask,
uint32_t mask_stride, int w, int h,
int subx, int suby, int bd) {
int subw, int subh, int bd) {
typedef void (*blend_fn)(
uint16_t * dst, uint32_t dst_stride, const uint16_t *src0,
uint32_t src0_stride, const uint16_t *src1, uint32_t src1_stride,
const uint8_t *mask, uint32_t mask_stride, int w, int h);
// Dimensions are: bd_index X width_index X subx X suby
// Dimensions are: bd_index X width_index X subw X subh
static const blend_fn blend[2][2][2][2] = {
{ // bd == 8 or 10
{ // w % 8 == 0
@ -858,18 +858,19 @@ void aom_highbd_blend_a64_mask_sse4_1(uint8_t *dst_8, uint32_t dst_stride,
assert(bd == 8 || bd == 10 || bd == 12);
if (UNLIKELY((h | w) & 3)) { // if (w <= 2 || h <= 2)
aom_highbd_blend_a64_mask_c(dst_8, dst_stride, src0_8, src0_stride, src1_8,
src1_stride, mask, mask_stride, w, h, subx,
suby, bd);
src1_stride, mask, mask_stride, w, h, subw,
subh, bd);
} else {
uint16_t *const dst = CONVERT_TO_SHORTPTR(dst_8);
const uint16_t *const src0 = CONVERT_TO_SHORTPTR(src0_8);
const uint16_t *const src1 = CONVERT_TO_SHORTPTR(src1_8);
blend[bd == 12][(w >> 2) & 1][subx != 0][suby != 0](
blend[bd == 12][(w >> 2) & 1][subw != 0][subh != 0](
dst, dst_stride, src0, src0_stride, src1, src1_stride, mask,
mask_stride, w, h);
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH
static INLINE void blend_a64_d16_mask_w16_sse41(
uint8_t *dst, const CONV_BUF_TYPE *src0, const CONV_BUF_TYPE *src1,
@ -1111,7 +1112,7 @@ void aom_lowbd_blend_a64_d16_mask_sse4_1(
//////////////////////////////////////////////////////////////////////////////
// aom_highbd_blend_a64_d16_mask_sse4_1()
//////////////////////////////////////////////////////////////////////////////
#if CONFIG_AV1_HIGHBITDEPTH
static INLINE void highbd_blend_a64_d16_mask_w4_sse4_1(
uint16_t *dst, int dst_stride, const CONV_BUF_TYPE *src0, int src0_stride,
const CONV_BUF_TYPE *src1, int src1_stride, const __m128i *mask0a,
@ -1556,3 +1557,4 @@ void aom_highbd_blend_a64_d16_mask_sse4_1(
subh, conv_params, bd);
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH

2
third_party/aom/aom_dsp/x86/blend_a64_vmask_sse4.c vendored
View file

@ -143,6 +143,7 @@ void aom_blend_a64_vmask_sse4_1(uint8_t *dst, uint32_t dst_stride,
h);
}
#if CONFIG_AV1_HIGHBITDEPTH
//////////////////////////////////////////////////////////////////////////////
// Implementation - No sub-sampling
//////////////////////////////////////////////////////////////////////////////
@ -281,3 +282,4 @@ void aom_highbd_blend_a64_vmask_sse4_1(
src1_stride, mask, w, h);
}
}
#endif // CONFIG_AV1_HIGHBITDEPTH

185
third_party/aom/aom_dsp/x86/blk_sse_sum_avx2.c vendored Normal file
View file

@ -0,0 +1,185 @@
/*
* Copyright (c) 2019, Alliance for Open Media. All rights reserved
*
* This source code is subject to the terms of the BSD 2 Clause License and
* the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
* was not distributed with this source code in the LICENSE file, you can
* obtain it at www.aomedia.org/license/software. If the Alliance for Open
* Media Patent License 1.0 was not distributed with this source code in the
* PATENTS file, you can obtain it at www.aomedia.org/license/patent.
*/
#include <immintrin.h>
#include "config/aom_dsp_rtcd.h"
static INLINE void accumulate_sse_sum(__m256i regx_sum, __m256i regx2_sum,
int *x_sum, int64_t *x2_sum) {
__m256i sum_buffer, sse_buffer;
__m128i out_buffer;
// Accumulate the various elements of register into first element.
sum_buffer = _mm256_permute2f128_si256(regx_sum, regx_sum, 1);
regx_sum = _mm256_add_epi32(sum_buffer, regx_sum);
regx_sum = _mm256_add_epi32(regx_sum, _mm256_srli_si256(regx_sum, 8));
regx_sum = _mm256_add_epi32(regx_sum, _mm256_srli_si256(regx_sum, 4));
sse_buffer = _mm256_permute2f128_si256(regx2_sum, regx2_sum, 1);
regx2_sum = _mm256_add_epi64(sse_buffer, regx2_sum);
regx2_sum = _mm256_add_epi64(regx2_sum, _mm256_srli_si256(regx2_sum, 8));
out_buffer = _mm256_castsi256_si128(regx_sum);
*x_sum += _mm_cvtsi128_si32(out_buffer);
out_buffer = _mm256_castsi256_si128(regx2_sum);
#if ARCH_X86_64
*x2_sum += _mm_cvtsi128_si64(out_buffer);
#else
{
int64_t tmp;
_mm_storel_epi64((__m128i *)&tmp, out_buffer);
*x2_sum += tmp;
}
#endif
}
static INLINE void sse_sum_wd4_avx2(const int16_t *data, int stride, int bh,
int *x_sum, int64_t *x2_sum) {
__m128i row1, row2, row3;
__m256i regx_sum, regx2_sum, load_pixels, sum_buffer, sse_buffer,
temp_buffer1, temp_buffer2, row_sum_buffer, row_sse_buffer;
const int16_t *data_tmp = data;
__m256i one = _mm256_set1_epi16(1);
regx_sum = _mm256_setzero_si256();
regx2_sum = regx_sum;
sum_buffer = _mm256_setzero_si256();
sse_buffer = sum_buffer;
for (int j = 0; j < (bh >> 2); ++j) {
// Load 4 rows at a time.
row1 = _mm_loadl_epi64((__m128i const *)(data_tmp));
row2 = _mm_loadl_epi64((__m128i const *)(data_tmp + stride));
row1 = _mm_unpacklo_epi64(row1, row2);
row2 = _mm_loadl_epi64((__m128i const *)(data_tmp + 2 * stride));
row3 = _mm_loadl_epi64((__m128i const *)(data_tmp + 3 * stride));
row2 = _mm_unpacklo_epi64(row2, row3);
load_pixels =
_mm256_insertf128_si256(_mm256_castsi128_si256(row1), row2, 1);
row_sum_buffer = _mm256_madd_epi16(load_pixels, one);
row_sse_buffer = _mm256_madd_epi16(load_pixels, load_pixels);
sum_buffer = _mm256_add_epi32(row_sum_buffer, sum_buffer);
sse_buffer = _mm256_add_epi32(row_sse_buffer, sse_buffer);
data_tmp += 4 * stride;
}
// To prevent 32-bit variable overflow, unpack the elements to 64-bit.
temp_buffer1 = _mm256_unpacklo_epi32(sse_buffer, _mm256_setzero_si256());
temp_buffer2 = _mm256_unpackhi_epi32(sse_buffer, _mm256_setzero_si256());
sse_buffer = _mm256_add_epi64(temp_buffer1, temp_buffer2);
regx_sum = _mm256_add_epi32(sum_buffer, regx_sum);
regx2_sum = _mm256_add_epi64(sse_buffer, regx2_sum);
accumulate_sse_sum(regx_sum, regx2_sum, x_sum, x2_sum);
}
static INLINE void sse_sum_wd8_avx2(const int16_t *data, int stride, int bh,
int *x_sum, int64_t *x2_sum) {
__m128i load_128bit, load_next_128bit;
__m256i regx_sum, regx2_sum, load_pixels, sum_buffer, sse_buffer,
temp_buffer1, temp_buffer2, row_sum_buffer, row_sse_buffer;
const int16_t *data_tmp = data;
__m256i one = _mm256_set1_epi16(1);
regx_sum = _mm256_setzero_si256();
regx2_sum = regx_sum;
sum_buffer = _mm256_setzero_si256();
sse_buffer = sum_buffer;
for (int j = 0; j < (bh >> 1); ++j) {
// Load 2 rows at a time.
load_128bit = _mm_loadu_si128((__m128i const *)(data_tmp));
load_next_128bit = _mm_loadu_si128((__m128i const *)(data_tmp + stride));
load_pixels = _mm256_insertf128_si256(_mm256_castsi128_si256(load_128bit),
load_next_128bit, 1);
row_sum_buffer = _mm256_madd_epi16(load_pixels, one);
row_sse_buffer = _mm256_madd_epi16(load_pixels, load_pixels);
sum_buffer = _mm256_add_epi32(row_sum_buffer, sum_buffer);
sse_buffer = _mm256_add_epi32(row_sse_buffer, sse_buffer);
data_tmp += 2 * stride;
}
temp_buffer1 = _mm256_unpacklo_epi32(sse_buffer, _mm256_setzero_si256());
temp_buffer2 = _mm256_unpackhi_epi32(sse_buffer, _mm256_setzero_si256());
sse_buffer = _mm256_add_epi64(temp_buffer1, temp_buffer2);
regx_sum = _mm256_add_epi32(sum_buffer, regx_sum);
regx2_sum = _mm256_add_epi64(sse_buffer, regx2_sum);
accumulate_sse_sum(regx_sum, regx2_sum, x_sum, x2_sum);
}
static INLINE void sse_sum_wd16_avx2(const int16_t *data, int stride, int bh,
int *x_sum, int64_t *x2_sum,
int loop_count) {
__m256i regx_sum, regx2_sum, load_pixels, sum_buffer, sse_buffer,
temp_buffer1, temp_buffer2, row_sum_buffer, row_sse_buffer;
const int16_t *data_tmp = data;
__m256i one = _mm256_set1_epi16(1);
regx_sum = _mm256_setzero_si256();
regx2_sum = regx_sum;
sum_buffer = _mm256_setzero_si256();
sse_buffer = sum_buffer;
for (int i = 0; i < loop_count; ++i) {
data_tmp = data + 16 * i;
for (int j = 0; j < bh; ++j) {
load_pixels = _mm256_lddqu_si256((__m256i const *)(data_tmp));
row_sum_buffer = _mm256_madd_epi16(load_pixels, one);
row_sse_buffer = _mm256_madd_epi16(load_pixels, load_pixels);
sum_buffer = _mm256_add_epi32(row_sum_buffer, sum_buffer);
sse_buffer = _mm256_add_epi32(row_sse_buffer, sse_buffer);
data_tmp += stride;
}
}
temp_buffer1 = _mm256_unpacklo_epi32(sse_buffer, _mm256_setzero_si256());
temp_buffer2 = _mm256_unpackhi_epi32(sse_buffer, _mm256_setzero_si256());
sse_buffer = _mm256_add_epi64(temp_buffer1, temp_buffer2);
regx_sum = _mm256_add_epi32(sum_buffer, regx_sum);
regx2_sum = _mm256_add_epi64(sse_buffer, regx2_sum);
accumulate_sse_sum(regx_sum, regx2_sum, x_sum, x2_sum);
}
void aom_get_blk_sse_sum_avx2(const int16_t *data, int stride, int bw, int bh,
int *x_sum, int64_t *x2_sum) {
*x_sum = 0;
*x2_sum = 0;
if ((bh & 3) == 0) {
switch (bw) {
// For smaller block widths, compute multiple rows simultaneously.
case 4: sse_sum_wd4_avx2(data, stride, bh, x_sum, x2_sum); break;
case 8: sse_sum_wd8_avx2(data, stride, bh, x_sum, x2_sum); break;
case 16:
case 32:
sse_sum_wd16_avx2(data, stride, bh, x_sum, x2_sum, bw >> 4);
break;
case 64:
// 32-bit variables will overflow for 64 rows at a single time, so
// compute 32 rows at a time.
if (bh <= 32) {
sse_sum_wd16_avx2(data, stride, bh, x_sum, x2_sum, bw >> 4);
} else {
sse_sum_wd16_avx2(data, stride, 32, x_sum, x2_sum, bw >> 4);
sse_sum_wd16_avx2(data + 32 * stride, stride, 32, x_sum, x2_sum,
bw >> 4);
}
break;
default: aom_get_blk_sse_sum_c(data, stride, bw, bh, x_sum, x2_sum);
}
} else {
aom_get_blk_sse_sum_c(data, stride, bw, bh, x_sum, x2_sum);
}
}

138
third_party/aom/aom_dsp/x86/blk_sse_sum_sse2.c vendored Normal file
View file

@ -0,0 +1,138 @@
/*
* Copyright (c) 2019, Alliance for Open Media. All rights reserved
*
* This source code is subject to the terms of the BSD 2 Clause License and
* the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
* was not distributed with this source code in the LICENSE file, you can
* obtain it at www.aomedia.org/license/software. If the Alliance for Open
* Media Patent License 1.0 was not distributed with this source code in the
* PATENTS file, you can obtain it at www.aomedia.org/license/patent.
*/
#include <emmintrin.h>
#include "config/aom_dsp_rtcd.h"
static INLINE void sse_sum_wd4_sse2(const int16_t *data, int stride, int bh,
int *x_sum, int64_t *x2_sum) {
const int16_t *data_tmp = data;
__m128i temp_buffer1, temp_buffer2;
__m128i load_pixels_low, load_pixels_hi, sum_buffer, sse_buffer;
__m128i one = _mm_set1_epi16(1);
__m128i regx_sum = _mm_setzero_si128();
__m128i regx2_sum = regx_sum;
for (int j = 0; j < (bh >> 1); ++j) {
// Load 2 rows (8 pixels) at a time.
load_pixels_low = _mm_loadl_epi64((__m128i const *)(data_tmp));
load_pixels_hi = _mm_loadl_epi64((__m128i const *)(data_tmp + stride));
load_pixels_low = _mm_unpacklo_epi64(load_pixels_low, load_pixels_hi);
sum_buffer = _mm_madd_epi16(load_pixels_low, one);
sse_buffer = _mm_madd_epi16(load_pixels_low, load_pixels_low);
regx_sum = _mm_add_epi32(sum_buffer, regx_sum);
regx2_sum = _mm_add_epi32(sse_buffer, regx2_sum);
data_tmp += 2 * stride;
}
regx_sum = _mm_add_epi32(regx_sum, _mm_srli_si128(regx_sum, 8));
regx_sum = _mm_add_epi32(regx_sum, _mm_srli_si128(regx_sum, 4));
*x_sum = _mm_cvtsi128_si32(regx_sum);
temp_buffer1 = _mm_unpacklo_epi32(regx2_sum, _mm_setzero_si128());
temp_buffer2 = _mm_unpackhi_epi32(regx2_sum, _mm_setzero_si128());
regx2_sum = _mm_add_epi64(temp_buffer1, temp_buffer2);
regx2_sum = _mm_add_epi64(regx2_sum, _mm_srli_si128(regx2_sum, 8));
#if ARCH_X86_64
*x2_sum += _mm_cvtsi128_si64(regx2_sum);
#else
{
int64_t tmp;
_mm_storel_epi64((__m128i *)&tmp, regx2_sum);
*x2_sum += tmp;
}
#endif
}
static INLINE void sse_sum_wd8_sse2(const int16_t *data, int stride, int bh,
int *x_sum, int64_t *x2_sum,
int loop_cycles) {
const int16_t *data_tmp;
__m128i temp_buffer1, temp_buffer2;
__m128i one = _mm_set1_epi16(1);
__m128i regx_sum = _mm_setzero_si128();
__m128i regx2_sum = regx_sum;
__m128i load_pixels, sum_buffer, sse_buffer;
for (int i = 0; i < loop_cycles; ++i) {
data_tmp = data + (8 * i);
for (int j = 0; j < bh; ++j) {
// Load 1 row (8-pixels) at a time.
load_pixels = _mm_loadu_si128((__m128i const *)(data_tmp));
sum_buffer = _mm_madd_epi16(load_pixels, one);
sse_buffer = _mm_madd_epi16(load_pixels, load_pixels);
regx_sum = _mm_add_epi32(sum_buffer, regx_sum);
regx2_sum = _mm_add_epi32(sse_buffer, regx2_sum);
data_tmp += stride;
}
}
regx_sum = _mm_add_epi32(regx_sum, _mm_srli_si128(regx_sum, 8));
regx_sum = _mm_add_epi32(regx_sum, _mm_srli_si128(regx_sum, 4));
*x_sum += _mm_cvtsi128_si32(regx_sum);
temp_buffer1 = _mm_unpacklo_epi32(regx2_sum, _mm_setzero_si128());
temp_buffer2 = _mm_unpackhi_epi32(regx2_sum, _mm_setzero_si128());
regx2_sum = _mm_add_epi64(temp_buffer1, temp_buffer2);
regx2_sum = _mm_add_epi64(regx2_sum, _mm_srli_si128(regx2_sum, 8));
#if ARCH_X86_64
*x2_sum += _mm_cvtsi128_si64(regx2_sum);
#else
{
int64_t tmp;
_mm_storel_epi64((__m128i *)&tmp, regx2_sum);
*x2_sum += tmp;
}
#endif
}
// This functions adds SSE2 Support for the functions 'get_blk_sse_sum_c'
void aom_get_blk_sse_sum_sse2(const int16_t *data, int stride, int bw, int bh,
int *x_sum, int64_t *x2_sum) {
*x_sum = 0;
*x2_sum = 0;
if ((bh & 3) == 0) {
switch (bw) {
case 4: sse_sum_wd4_sse2(data, stride, bh, x_sum, x2_sum); break;
case 8:
case 16:
sse_sum_wd8_sse2(data, stride, bh, x_sum, x2_sum, bw >> 3);
break;
// For widths 32 and 64, the registers may overflow. So compute
// partial widths at a time.
case 32:
if (bh <= 32) {
sse_sum_wd8_sse2(data, stride, bh, x_sum, x2_sum, bw >> 3);
break;
} else {
sse_sum_wd8_sse2(data, stride, 32, x_sum, x2_sum, bw >> 3);
sse_sum_wd8_sse2(data + 32 * stride, stride, 32, x_sum, x2_sum,
bw >> 3);
break;
}
case 64:
if (bh <= 16) {
sse_sum_wd8_sse2(data, stride, bh, x_sum, x2_sum, bw >> 3);
break;
} else {
for (int i = 0; i < bh; i += 16)
sse_sum_wd8_sse2(data + i * stride, stride, 16, x_sum, x2_sum,
bw >> 3);
break;
}
default: aom_get_blk_sse_sum_c(data, stride, bw, bh, x_sum, x2_sum);
}
} else {
aom_get_blk_sse_sum_c(data, stride, bw, bh, x_sum, x2_sum);
}
}

2
third_party/aom/aom_dsp/x86/convolve.h vendored
View file

@ -107,6 +107,7 @@ typedef void filter8_1dfunction(const uint8_t *src_ptr, ptrdiff_t src_pitch,
} \
}
#if CONFIG_AV1_HIGHBITDEPTH
typedef void highbd_filter8_1dfunction(const uint16_t *src_ptr,
const ptrdiff_t src_pitch,
uint16_t *output_ptr,
@ -197,5 +198,6 @@ typedef void highbd_filter8_1dfunction(const uint16_t *src_ptr,
dst_stride, filter_x, x_step_q4, filter_y, y_step_q4, w, h, bd); \
} \
}
#endif // CONFIG_AV1_HIGHBITDEPTH
#endif // AOM_AOM_DSP_X86_CONVOLVE_H_

6
third_party/aom/aom_dsp/x86/convolve_avx2.h vendored
View file

@ -138,8 +138,8 @@ DECLARE_ALIGNED(32, static const uint8_t, filt4_global_avx2[32]) = {
xx_storel_32(p_0, res_0); \
xx_storel_32(p_1, res_1); \
} else { \
*(uint16_t *)p_0 = _mm_cvtsi128_si32(res_0); \
*(uint16_t *)p_1 = _mm_cvtsi128_si32(res_1); \
*(uint16_t *)p_0 = (uint16_t)_mm_cvtsi128_si32(res_0); \
*(uint16_t *)p_1 = (uint16_t)_mm_cvtsi128_si32(res_1); \
} \
\
s[0] = s[1]; \
@ -281,7 +281,7 @@ static INLINE void prepare_coeffs_lowbd(
// Since all filter co-efficients are even, this change will not affect the
// end result
assert(_mm_test_all_zeros(_mm_and_si128(coeffs_8, _mm_set1_epi16(1)),
_mm_set1_epi16(0xffff)));
_mm_set1_epi16((short)0xffff)));
const __m256i coeffs_1 = _mm256_srai_epi16(filter_coeffs, 1);

4
third_party/aom/aom_dsp/x86/fwd_txfm_sse2.c vendored
View file

@ -22,8 +22,12 @@
#include "aom_dsp/x86/fwd_txfm_impl_sse2.h"
#undef FDCT8x8_2D
#if CONFIG_AV1_HIGHBITDEPTH
#undef DCT_HIGH_BIT_DEPTH
#define DCT_HIGH_BIT_DEPTH 1
#define FDCT8x8_2D aom_highbd_fdct8x8_sse2
#include "aom_dsp/x86/fwd_txfm_impl_sse2.h" // NOLINT
#undef FDCT8x8_2D
#endif

457
third_party/aom/aom_dsp/x86/highbd_adaptive_quantize_avx2.c vendored Normal file
View file

@ -0,0 +1,457 @@
/*
* Copyright (c) 2019, Alliance for Open Media. All rights reserved
*
* This source code is subject to the terms of the BSD 2 Clause License and
* the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
* was not distributed with this source code in the LICENSE file, you can
* obtain it at www.aomedia.org/license/software. If the Alliance for Open
* Media Patent License 1.0 was not distributed with this source code in the
* PATENTS file, you can obtain it at www.aomedia.org/license/patent.
*/
#include <immintrin.h>
#include "config/aom_dsp_rtcd.h"
#include "aom/aom_integer.h"
#include "aom_dsp/x86/quantize_x86.h"
#include "av1/encoder/av1_quantize.h"
static INLINE void highbd_load_b_values_avx2(
const int16_t *zbin_ptr, __m256i *zbin, const int16_t *round_ptr,
__m256i *round, const int16_t *quant_ptr, __m256i *quant,
const int16_t *dequant_ptr, __m256i *dequant, const int16_t *shift_ptr,
__m256i *shift) {
*zbin = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)zbin_ptr));
*zbin = _mm256_sub_epi32(*zbin, _mm256_set1_epi32(1));
*round = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)round_ptr));
*quant = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)quant_ptr));
*dequant =
_mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)dequant_ptr));
*shift = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)shift_ptr));
}
static INLINE void highbd_update_mask1_avx2(__m256i *cmp_mask,
const int16_t *iscan_ptr,
int *is_found, __m256i *mask) {
__m256i temp_mask = _mm256_setzero_si256();
if (_mm256_movemask_epi8(*cmp_mask)) {
__m256i iscan = _mm256_loadu_si256((const __m256i *)(iscan_ptr));
temp_mask = _mm256_and_si256(*cmp_mask, iscan);
*is_found = 1;
}
*mask = _mm256_max_epi16(temp_mask, *mask);
}
static INLINE void highbd_update_mask0_avx2(__m256i *qcoeff0, __m256i *qcoeff1,
__m256i *threshold,
const int16_t *iscan_ptr,
int *is_found, __m256i *mask) {
__m256i coeff[2], cmp_mask0, cmp_mask1;
coeff[0] = _mm256_slli_epi32(*qcoeff0, AOM_QM_BITS);
cmp_mask0 = _mm256_cmpgt_epi32(coeff[0], threshold[0]);
coeff[1] = _mm256_slli_epi32(*qcoeff1, AOM_QM_BITS);
cmp_mask1 = _mm256_cmpgt_epi32(coeff[1], threshold[1]);
cmp_mask0 =
_mm256_permute4x64_epi64(_mm256_packs_epi32(cmp_mask0, cmp_mask1), 0xd8);
highbd_update_mask1_avx2(&cmp_mask0, iscan_ptr, is_found, mask);
}
static INLINE void highbd_mul_shift_avx2(const __m256i *x, const __m256i *y,
__m256i *p, const int shift) {
__m256i prod_lo = _mm256_mul_epi32(*x, *y);
__m256i prod_hi = _mm256_srli_epi64(*x, 32);
const __m256i mult_hi = _mm256_srli_epi64(*y, 32);
prod_hi = _mm256_mul_epi32(prod_hi, mult_hi);
prod_lo = _mm256_srli_epi64(prod_lo, shift);
prod_hi = _mm256_srli_epi64(prod_hi, shift);
prod_hi = _mm256_slli_epi64(prod_hi, 32);
*p = _mm256_blend_epi32(prod_lo, prod_hi, 0xaa);
}
static INLINE void highbd_calculate_qcoeff_avx2(__m256i *coeff,
const __m256i *round,
const __m256i *quant,
const __m256i *shift,
const int *log_scale) {
__m256i tmp, qcoeff;
qcoeff = _mm256_add_epi32(*coeff, *round);
highbd_mul_shift_avx2(&qcoeff, quant, &tmp, 16);
qcoeff = _mm256_add_epi32(tmp, qcoeff);
highbd_mul_shift_avx2(&qcoeff, shift, coeff, 16 - *log_scale);
}
static INLINE __m256i highbd_calculate_dqcoeff_avx2(__m256i qcoeff,
__m256i dequant) {
return _mm256_mullo_epi32(qcoeff, dequant);
}
static INLINE __m256i highbd_calculate_dqcoeff_log_scale_avx2(
__m256i qcoeff, __m256i dequant, const int log_scale) {
__m256i abs_coeff = _mm256_abs_epi32(qcoeff);
highbd_mul_shift_avx2(&abs_coeff, &dequant, &abs_coeff, log_scale);
return _mm256_sign_epi32(abs_coeff, qcoeff);
}
static INLINE void highbd_store_coefficients_avx2(__m256i coeff0,
__m256i coeff1,
tran_low_t *coeff_ptr) {
_mm256_store_si256((__m256i *)(coeff_ptr), coeff0);
_mm256_store_si256((__m256i *)(coeff_ptr + 8), coeff1);
}
void aom_highbd_quantize_b_adaptive_avx2(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
int index = 16;
int non_zero_count = 0;
int non_zero_count_prescan_add_zero = 0;
int is_found0 = 0, is_found1 = 0;
int eob = -1;
const __m256i zero = _mm256_setzero_si256();
__m256i zbin, round, quant, dequant, shift;
__m256i coeff0, qcoeff0, coeff1, qcoeff1;
__m256i cmp_mask, mask0 = zero, mask1 = zero;
__m128i temp_mask0, temp_mask1;
int prescan_add[2];
int thresh[2];
const int log_scale = 0;
const qm_val_t wt = (1 << AOM_QM_BITS);
for (int i = 0; i < 2; ++i) {
prescan_add[i] = ROUND_POWER_OF_TWO(dequant_ptr[i] * EOB_FACTOR, 7);
thresh[i] = (zbin_ptr[i] * wt + prescan_add[i]) - 1;
}
__m256i threshold[2];
threshold[0] = _mm256_set1_epi32(thresh[0]);
threshold[1] = _mm256_set1_epi32(thresh[1]);
threshold[0] = _mm256_blend_epi32(threshold[0], threshold[1], 0xfe);
#if SKIP_EOB_FACTOR_ADJUST
int first = -1;
#endif
// Setup global values.
highbd_load_b_values_avx2(zbin_ptr, &zbin, round_ptr, &round, quant_ptr,
&quant, dequant_ptr, &dequant, quant_shift_ptr,
&shift);
// Do DC and first 15 AC.
coeff0 = _mm256_load_si256((__m256i *)(coeff_ptr));
qcoeff0 = _mm256_abs_epi32(coeff0);
coeff1 = _mm256_load_si256((__m256i *)(coeff_ptr + 8));
qcoeff1 = _mm256_abs_epi32(coeff1);
highbd_update_mask0_avx2(&qcoeff0, &qcoeff1, threshold, iscan, &is_found0,
&mask0);
__m256i temp0 = _mm256_cmpgt_epi32(qcoeff0, zbin);
zbin = _mm256_unpackhi_epi64(zbin, zbin);
__m256i temp1 = _mm256_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm256_permute4x64_epi64(_mm256_packs_epi32(temp0, temp1), 0xd8);
highbd_update_mask1_avx2(&cmp_mask, iscan, &is_found1, &mask1);
threshold[0] = threshold[1];
if (_mm256_movemask_epi8(cmp_mask) == 0) {
_mm256_store_si256((__m256i *)(qcoeff_ptr), zero);
_mm256_store_si256((__m256i *)(qcoeff_ptr + 8), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr + 8), zero);
round = _mm256_unpackhi_epi64(round, round);
quant = _mm256_unpackhi_epi64(quant, quant);
shift = _mm256_unpackhi_epi64(shift, shift);
dequant = _mm256_unpackhi_epi64(dequant, dequant);
} else {
highbd_calculate_qcoeff_avx2(&qcoeff0, &round, &quant, &shift, &log_scale);
round = _mm256_unpackhi_epi64(round, round);
quant = _mm256_unpackhi_epi64(quant, quant);
shift = _mm256_unpackhi_epi64(shift, shift);
highbd_calculate_qcoeff_avx2(&qcoeff1, &round, &quant, &shift, &log_scale);
// Reinsert signs
qcoeff0 = _mm256_sign_epi32(qcoeff0, coeff0);
qcoeff1 = _mm256_sign_epi32(qcoeff1, coeff1);
// Mask out zbin threshold coeffs
qcoeff0 = _mm256_and_si256(qcoeff0, temp0);
qcoeff1 = _mm256_and_si256(qcoeff1, temp1);
highbd_store_coefficients_avx2(qcoeff0, qcoeff1, qcoeff_ptr);
coeff0 = highbd_calculate_dqcoeff_avx2(qcoeff0, dequant);
dequant = _mm256_unpackhi_epi64(dequant, dequant);
coeff1 = highbd_calculate_dqcoeff_avx2(qcoeff1, dequant);
highbd_store_coefficients_avx2(coeff0, coeff1, dqcoeff_ptr);
}
// AC only loop.
while (index < n_coeffs) {
coeff0 = _mm256_load_si256((__m256i *)(coeff_ptr + index));
qcoeff0 = _mm256_abs_epi32(coeff0);
coeff1 = _mm256_load_si256((__m256i *)(coeff_ptr + index + 8));
qcoeff1 = _mm256_abs_epi32(coeff1);
highbd_update_mask0_avx2(&qcoeff0, &qcoeff1, threshold, iscan + index,
&is_found0, &mask0);
temp0 = _mm256_cmpgt_epi32(qcoeff0, zbin);
temp1 = _mm256_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm256_permute4x64_epi64(_mm256_packs_epi32(temp0, temp1), 0xd8);
highbd_update_mask1_avx2(&cmp_mask, iscan + index, &is_found1, &mask1);
if (_mm256_movemask_epi8(cmp_mask) == 0) {
_mm256_store_si256((__m256i *)(qcoeff_ptr + index), zero);
_mm256_store_si256((__m256i *)(qcoeff_ptr + index + 8), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr + index), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr + index + 8), zero);
index += 16;
continue;
}
highbd_calculate_qcoeff_avx2(&qcoeff0, &round, &quant, &shift, &log_scale);
highbd_calculate_qcoeff_avx2(&qcoeff1, &round, &quant, &shift, &log_scale);
qcoeff0 = _mm256_sign_epi32(qcoeff0, coeff0);
qcoeff1 = _mm256_sign_epi32(qcoeff1, coeff1);
qcoeff0 = _mm256_and_si256(qcoeff0, temp0);
qcoeff1 = _mm256_and_si256(qcoeff1, temp1);
highbd_store_coefficients_avx2(qcoeff0, qcoeff1, qcoeff_ptr + index);
coeff0 = highbd_calculate_dqcoeff_avx2(qcoeff0, dequant);
coeff1 = highbd_calculate_dqcoeff_avx2(qcoeff1, dequant);
highbd_store_coefficients_avx2(coeff0, coeff1, dqcoeff_ptr + index);
index += 16;
}
if (is_found0) {
temp_mask0 = _mm_max_epi16(_mm256_castsi256_si128(mask0),
_mm256_extracti128_si256(mask0, 1));
non_zero_count = calculate_non_zero_count(temp_mask0);
}
if (is_found1) {
temp_mask1 = _mm_max_epi16(_mm256_castsi256_si128(mask1),
_mm256_extracti128_si256(mask1, 1));
non_zero_count_prescan_add_zero = calculate_non_zero_count(temp_mask1);
}
for (int i = non_zero_count_prescan_add_zero - 1; i >= non_zero_count; i--) {
const int rc = scan[i];
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
}
for (int i = non_zero_count - 1; i >= 0; i--) {
const int rc = scan[i];
if (qcoeff_ptr[rc]) {
eob = i;
break;
}
}
*eob_ptr = eob + 1;
#if SKIP_EOB_FACTOR_ADJUST
// TODO(Aniket): Experiment the following loop with intrinsic by combining
// with the quantization loop above
for (int i = 0; i < non_zero_count; i++) {
const int rc = scan[i];
const int qcoeff = qcoeff_ptr[rc];
if (qcoeff) {
first = i;
break;
}
}
if ((*eob_ptr - 1) >= 0 && first == (*eob_ptr - 1)) {
const int rc = scan[(*eob_ptr - 1)];
if (qcoeff_ptr[rc] == 1 || qcoeff_ptr[rc] == -1) {
const int coeff = coeff_ptr[rc] * wt;
const int coeff_sign = (coeff >> 31);
const int abs_coeff = (coeff ^ coeff_sign) - coeff_sign;
const int factor = EOB_FACTOR + SKIP_EOB_FACTOR_ADJUST;
const int prescan_add_val =
ROUND_POWER_OF_TWO(dequant_ptr[rc != 0] * factor, 7);
if (abs_coeff <
(zbin_ptr[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val)) {
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
*eob_ptr = 0;
}
}
}
#endif
}
void aom_highbd_quantize_b_32x32_adaptive_avx2(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
int index = 16;
int non_zero_count = 0;
int non_zero_count_prescan_add_zero = 0;
int is_found0 = 0, is_found1 = 0;
int eob = -1;
const int log_scale = 1;
const __m256i zero = _mm256_setzero_si256();
__m256i zbin, round, quant, dequant, shift;
__m256i coeff0, qcoeff0, coeff1, qcoeff1;
__m256i cmp_mask, mask0 = zero, mask1 = zero;
__m128i temp_mask0, temp_mask1;
const __m256i one = _mm256_set1_epi32(1);
const __m256i log_scale_vec = _mm256_set1_epi32(log_scale);
int prescan_add[2];
int thresh[2];
const int zbins[2] = { ROUND_POWER_OF_TWO(zbin_ptr[0], log_scale),
ROUND_POWER_OF_TWO(zbin_ptr[1], log_scale) };
const qm_val_t wt = (1 << AOM_QM_BITS);
for (int i = 0; i < 2; ++i) {
prescan_add[i] = ROUND_POWER_OF_TWO(dequant_ptr[i] * EOB_FACTOR, 7);
thresh[i] = (zbins[i] * wt + prescan_add[i]) - 1;
}
__m256i threshold[2];
threshold[0] = _mm256_set1_epi32(thresh[0]);
threshold[1] = _mm256_set1_epi32(thresh[1]);
threshold[0] = _mm256_blend_epi32(threshold[0], threshold[1], 0xfe);
#if SKIP_EOB_FACTOR_ADJUST
int first = -1;
#endif
// Setup global values.
zbin = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)zbin_ptr));
round = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)round_ptr));
quant = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)quant_ptr));
dequant = _mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)dequant_ptr));
shift =
_mm256_cvtepi16_epi32(_mm_load_si128((const __m128i *)quant_shift_ptr));
// Shift with rounding.
zbin = _mm256_add_epi32(zbin, log_scale_vec);
round = _mm256_add_epi32(round, log_scale_vec);
zbin = _mm256_srli_epi32(zbin, log_scale);
round = _mm256_srli_epi32(round, log_scale);
zbin = _mm256_sub_epi32(zbin, one);
// Do DC and first 15 AC.
coeff0 = _mm256_load_si256((__m256i *)(coeff_ptr));
qcoeff0 = _mm256_abs_epi32(coeff0);
coeff1 = _mm256_load_si256((__m256i *)(coeff_ptr + 8));
qcoeff1 = _mm256_abs_epi32(coeff1);
highbd_update_mask0_avx2(&qcoeff0, &qcoeff1, threshold, iscan, &is_found0,
&mask0);
__m256i temp0 = _mm256_cmpgt_epi32(qcoeff0, zbin);
zbin = _mm256_permute2x128_si256(zbin, zbin, 0x11);
__m256i temp1 = _mm256_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm256_permute4x64_epi64(_mm256_packs_epi32(temp0, temp1), 0xd8);
highbd_update_mask1_avx2(&cmp_mask, iscan, &is_found1, &mask1);
threshold[0] = threshold[1];
if (_mm256_movemask_epi8(cmp_mask) == 0) {
_mm256_store_si256((__m256i *)(qcoeff_ptr), zero);
_mm256_store_si256((__m256i *)(qcoeff_ptr + 8), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr + 8), zero);
round = _mm256_permute2x128_si256(round, round, 0x11);
quant = _mm256_permute2x128_si256(quant, quant, 0x11);
shift = _mm256_permute2x128_si256(shift, shift, 0x11);
dequant = _mm256_permute2x128_si256(dequant, dequant, 0x11);
} else {
highbd_calculate_qcoeff_avx2(&qcoeff0, &round, &quant, &shift, &log_scale);
round = _mm256_permute2x128_si256(round, round, 0x11);
quant = _mm256_permute2x128_si256(quant, quant, 0x11);
shift = _mm256_permute2x128_si256(shift, shift, 0x11);
highbd_calculate_qcoeff_avx2(&qcoeff1, &round, &quant, &shift, &log_scale);
// Reinsert signs
qcoeff0 = _mm256_sign_epi32(qcoeff0, coeff0);
qcoeff1 = _mm256_sign_epi32(qcoeff1, coeff1);
// Mask out zbin threshold coeffs
qcoeff0 = _mm256_and_si256(qcoeff0, temp0);
qcoeff1 = _mm256_and_si256(qcoeff1, temp1);
highbd_store_coefficients_avx2(qcoeff0, qcoeff1, qcoeff_ptr);
coeff0 =
highbd_calculate_dqcoeff_log_scale_avx2(qcoeff0, dequant, log_scale);
dequant = _mm256_permute2x128_si256(dequant, dequant, 0x11);
coeff1 =
highbd_calculate_dqcoeff_log_scale_avx2(qcoeff1, dequant, log_scale);
highbd_store_coefficients_avx2(coeff0, coeff1, dqcoeff_ptr);
}
// AC only loop.
while (index < n_coeffs) {
coeff0 = _mm256_load_si256((__m256i *)(coeff_ptr + index));
qcoeff0 = _mm256_abs_epi32(coeff0);
coeff1 = _mm256_load_si256((__m256i *)(coeff_ptr + index + 8));
qcoeff1 = _mm256_abs_epi32(coeff1);
highbd_update_mask0_avx2(&qcoeff0, &qcoeff1, threshold, iscan + index,
&is_found0, &mask0);
temp0 = _mm256_cmpgt_epi32(qcoeff0, zbin);
temp1 = _mm256_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm256_permute4x64_epi64(_mm256_packs_epi32(temp0, temp1), 0xd8);
highbd_update_mask1_avx2(&cmp_mask, iscan + index, &is_found1, &mask1);
if (_mm256_movemask_epi8(cmp_mask) == 0) {
_mm256_store_si256((__m256i *)(qcoeff_ptr + index), zero);
_mm256_store_si256((__m256i *)(qcoeff_ptr + index + 8), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr + index), zero);
_mm256_store_si256((__m256i *)(dqcoeff_ptr + index + 8), zero);
index += 16;
continue;
}
highbd_calculate_qcoeff_avx2(&qcoeff0, &round, &quant, &shift, &log_scale);
highbd_calculate_qcoeff_avx2(&qcoeff1, &round, &quant, &shift, &log_scale);
qcoeff0 = _mm256_sign_epi32(qcoeff0, coeff0);
qcoeff1 = _mm256_sign_epi32(qcoeff1, coeff1);
qcoeff0 = _mm256_and_si256(qcoeff0, temp0);
qcoeff1 = _mm256_and_si256(qcoeff1, temp1);
highbd_store_coefficients_avx2(qcoeff0, qcoeff1, qcoeff_ptr + index);
coeff0 =
highbd_calculate_dqcoeff_log_scale_avx2(qcoeff0, dequant, log_scale);
coeff1 =
highbd_calculate_dqcoeff_log_scale_avx2(qcoeff1, dequant, log_scale);
highbd_store_coefficients_avx2(coeff0, coeff1, dqcoeff_ptr + index);
index += 16;
}
if (is_found0) {
temp_mask0 = _mm_max_epi16(_mm256_castsi256_si128(mask0),
_mm256_extracti128_si256(mask0, 1));
non_zero_count = calculate_non_zero_count(temp_mask0);
}
if (is_found1) {
temp_mask1 = _mm_max_epi16(_mm256_castsi256_si128(mask1),
_mm256_extracti128_si256(mask1, 1));
non_zero_count_prescan_add_zero = calculate_non_zero_count(temp_mask1);
}
for (int i = non_zero_count_prescan_add_zero - 1; i >= non_zero_count; i--) {
const int rc = scan[i];
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
}
for (int i = non_zero_count - 1; i >= 0; i--) {
const int rc = scan[i];
if (qcoeff_ptr[rc]) {
eob = i;
break;
}
}
*eob_ptr = eob + 1;
#if SKIP_EOB_FACTOR_ADJUST
// TODO(Aniket): Experiment the following loop with intrinsic by combining
// with the quantization loop above
for (int i = 0; i < non_zero_count; i++) {
const int rc = scan[i];
const int qcoeff = qcoeff_ptr[rc];
if (qcoeff) {
first = i;
break;
}
}
if ((*eob_ptr - 1) >= 0 && first == (*eob_ptr - 1)) {
const int rc = scan[(*eob_ptr - 1)];
if (qcoeff_ptr[rc] == 1 || qcoeff_ptr[rc] == -1) {
const int coeff = coeff_ptr[rc] * wt;
const int coeff_sign = (coeff >> 31);
const int abs_coeff = (coeff ^ coeff_sign) - coeff_sign;
const int factor = EOB_FACTOR + SKIP_EOB_FACTOR_ADJUST;
const int prescan_add_val =
ROUND_POWER_OF_TWO(dequant_ptr[rc != 0] * factor, 7);
if (abs_coeff < (zbins[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val)) {
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
*eob_ptr = 0;
}
}
}
#endif
}

732
third_party/aom/aom_dsp/x86/highbd_adaptive_quantize_sse2.c vendored Normal file
View file

@ -0,0 +1,732 @@
/*
* Copyright (c) 2019, Alliance for Open Media. All rights reserved
*
* This source code is subject to the terms of the BSD 2 Clause License and
* the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
* was not distributed with this source code in the LICENSE file, you can
* obtain it at www.aomedia.org/license/software. If the Alliance for Open
* Media Patent License 1.0 was not distributed with this source code in the
* PATENTS file, you can obtain it at www.aomedia.org/license/patent.
*/
#include <emmintrin.h>
#include "config/aom_dsp_rtcd.h"
#include "aom/aom_integer.h"
#include "aom_dsp/x86/quantize_x86.h"
#include "av1/encoder/av1_quantize.h"
static INLINE __m128i highbd_invert_sign_64bit_sse2(__m128i a, __m128i sign) {
a = _mm_xor_si128(a, sign);
return _mm_sub_epi64(a, sign);
}
static INLINE void highbd_mul_shift_sse2(const __m128i *x, const __m128i *y,
__m128i *p, const int shift) {
__m128i sign = _mm_srai_epi32(*y, 31);
__m128i sign_lo = _mm_unpacklo_epi32(sign, sign);
__m128i sign_hi = _mm_unpackhi_epi32(sign, sign);
__m128i abs_y = invert_sign_32_sse2(*y, sign);
__m128i prod_lo = _mm_mul_epu32(*x, abs_y);
__m128i prod_hi = _mm_srli_epi64(*x, 32);
const __m128i mult_hi = _mm_srli_epi64(abs_y, 32);
prod_hi = _mm_mul_epu32(prod_hi, mult_hi);
prod_lo = highbd_invert_sign_64bit_sse2(prod_lo, sign_lo);
prod_hi = highbd_invert_sign_64bit_sse2(prod_hi, sign_hi);
prod_lo = _mm_srli_epi64(prod_lo, shift);
const __m128i mask = _mm_set_epi32(0, -1, 0, -1);
prod_lo = _mm_and_si128(prod_lo, mask);
prod_hi = _mm_srli_epi64(prod_hi, shift);
prod_hi = _mm_slli_epi64(prod_hi, 32);
*p = _mm_or_si128(prod_lo, prod_hi);
}
static INLINE void highbd_calculate_qcoeff(__m128i *coeff, const __m128i *round,
const __m128i *quant,
const __m128i *shift,
const int *log_scale) {
__m128i tmp, qcoeff;
qcoeff = _mm_add_epi32(*coeff, *round);
highbd_mul_shift_sse2(&qcoeff, quant, &tmp, 16);
qcoeff = _mm_add_epi32(tmp, qcoeff);
highbd_mul_shift_sse2(&qcoeff, shift, coeff, 16 - *log_scale);
}
static INLINE void highbd_update_mask1(__m128i *cmp_mask0,
const int16_t *iscan_ptr, int *is_found,
__m128i *mask) {
__m128i temp_mask = _mm_setzero_si128();
if (_mm_movemask_epi8(*cmp_mask0)) {
__m128i iscan0 = _mm_load_si128((const __m128i *)(iscan_ptr));
__m128i mask0 = _mm_and_si128(*cmp_mask0, iscan0);
temp_mask = mask0;
*is_found = 1;
}
*mask = _mm_max_epi16(temp_mask, *mask);
}
static INLINE void highbd_update_mask0(__m128i *qcoeff0, __m128i *qcoeff1,
__m128i *threshold,
const int16_t *iscan_ptr, int *is_found,
__m128i *mask) {
__m128i coeff[2], cmp_mask0, cmp_mask1;
coeff[0] = _mm_slli_epi32(*qcoeff0, AOM_QM_BITS);
cmp_mask0 = _mm_cmpgt_epi32(coeff[0], threshold[0]);
coeff[1] = _mm_slli_epi32(*qcoeff1, AOM_QM_BITS);
cmp_mask1 = _mm_cmpgt_epi32(coeff[1], threshold[1]);
cmp_mask0 = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask0, iscan_ptr, is_found, mask);
}
static INLINE __m128i highbd_calculate_dqcoeff(__m128i qcoeff, __m128i dequant,
const int log_scale) {
__m128i coeff_sign = _mm_srai_epi32(qcoeff, 31);
__m128i abs_coeff = invert_sign_32_sse2(qcoeff, coeff_sign);
highbd_mul_shift_sse2(&abs_coeff, &dequant, &abs_coeff, log_scale);
return invert_sign_32_sse2(abs_coeff, coeff_sign);
}
void aom_highbd_quantize_b_adaptive_sse2(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
int index = 8;
const int log_scale = 0;
int non_zero_count = 0;
int non_zero_count_prescan_add_zero = 0;
int is_found0 = 0, is_found1 = 0;
int eob = -1;
const __m128i zero = _mm_setzero_si128();
const __m128i one = _mm_set1_epi32(1);
__m128i zbin, round, quant, dequant, shift;
__m128i coeff0, coeff1, coeff0_sign, coeff1_sign;
__m128i qcoeff0, qcoeff1;
__m128i cmp_mask0, cmp_mask1, cmp_mask;
__m128i all_zero;
__m128i mask0 = zero, mask1 = zero;
int prescan_add[2];
int thresh[4];
const qm_val_t wt = (1 << AOM_QM_BITS);
for (int i = 0; i < 2; ++i) {
prescan_add[i] = ROUND_POWER_OF_TWO(dequant_ptr[i] * EOB_FACTOR, 7);
thresh[i] = (zbin_ptr[i] * wt + prescan_add[i]) - 1;
}
thresh[2] = thresh[3] = thresh[1];
__m128i threshold[2];
threshold[0] = _mm_loadu_si128((__m128i *)&thresh[0]);
threshold[1] = _mm_unpackhi_epi64(threshold[0], threshold[0]);
#if SKIP_EOB_FACTOR_ADJUST
int first = -1;
#endif
// Setup global values.
zbin = _mm_load_si128((const __m128i *)zbin_ptr);
round = _mm_load_si128((const __m128i *)round_ptr);
quant = _mm_load_si128((const __m128i *)quant_ptr);
dequant = _mm_load_si128((const __m128i *)dequant_ptr);
shift = _mm_load_si128((const __m128i *)quant_shift_ptr);
__m128i zbin_sign = _mm_srai_epi16(zbin, 15);
__m128i round_sign = _mm_srai_epi16(round, 15);
__m128i quant_sign = _mm_srai_epi16(quant, 15);
__m128i dequant_sign = _mm_srai_epi16(dequant, 15);
__m128i shift_sign = _mm_srai_epi16(shift, 15);
zbin = _mm_unpacklo_epi16(zbin, zbin_sign);
round = _mm_unpacklo_epi16(round, round_sign);
quant = _mm_unpacklo_epi16(quant, quant_sign);
dequant = _mm_unpacklo_epi16(dequant, dequant_sign);
shift = _mm_unpacklo_epi16(shift, shift_sign);
zbin = _mm_sub_epi32(zbin, one);
// Do DC and first 15 AC.
coeff0 = _mm_load_si128((__m128i *)(coeff_ptr));
coeff1 = _mm_load_si128((__m128i *)(coeff_ptr + 4));
coeff0_sign = _mm_srai_epi32(coeff0, 31);
coeff1_sign = _mm_srai_epi32(coeff1, 31);
qcoeff0 = invert_sign_32_sse2(coeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(coeff1, coeff1_sign);
highbd_update_mask0(&qcoeff0, &qcoeff1, threshold, iscan, &is_found0, &mask0);
cmp_mask0 = _mm_cmpgt_epi32(qcoeff0, zbin);
zbin = _mm_unpackhi_epi64(zbin, zbin); // Switch DC to AC
cmp_mask1 = _mm_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask, iscan, &is_found1, &mask1);
threshold[0] = threshold[1];
all_zero = _mm_or_si128(cmp_mask0, cmp_mask1);
if (_mm_movemask_epi8(all_zero) == 0) {
_mm_store_si128((__m128i *)(qcoeff_ptr), zero);
_mm_store_si128((__m128i *)(qcoeff_ptr + 4), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + 4), zero);
round = _mm_unpackhi_epi64(round, round);
quant = _mm_unpackhi_epi64(quant, quant);
shift = _mm_unpackhi_epi64(shift, shift);
dequant = _mm_unpackhi_epi64(dequant, dequant);
} else {
highbd_calculate_qcoeff(&qcoeff0, &round, &quant, &shift, &log_scale);
round = _mm_unpackhi_epi64(round, round);
quant = _mm_unpackhi_epi64(quant, quant);
shift = _mm_unpackhi_epi64(shift, shift);
highbd_calculate_qcoeff(&qcoeff1, &round, &quant, &shift, &log_scale);
// Reinsert signs
qcoeff0 = invert_sign_32_sse2(qcoeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(qcoeff1, coeff1_sign);
// Mask out zbin threshold coeffs
qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
_mm_store_si128((__m128i *)(qcoeff_ptr), qcoeff0);
_mm_store_si128((__m128i *)(qcoeff_ptr + 4), qcoeff1);
coeff0 = highbd_calculate_dqcoeff(qcoeff0, dequant, log_scale);
dequant = _mm_unpackhi_epi64(dequant, dequant);
coeff1 = highbd_calculate_dqcoeff(qcoeff1, dequant, log_scale);
_mm_store_si128((__m128i *)(dqcoeff_ptr), coeff0);
_mm_store_si128((__m128i *)(dqcoeff_ptr + 4), coeff1);
}
// AC only loop.
while (index < n_coeffs) {
coeff0 = _mm_load_si128((__m128i *)(coeff_ptr + index));
coeff1 = _mm_load_si128((__m128i *)(coeff_ptr + index + 4));
coeff0_sign = _mm_srai_epi32(coeff0, 31);
coeff1_sign = _mm_srai_epi32(coeff1, 31);
qcoeff0 = invert_sign_32_sse2(coeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(coeff1, coeff1_sign);
highbd_update_mask0(&qcoeff0, &qcoeff1, threshold, iscan + index,
&is_found0, &mask0);
cmp_mask0 = _mm_cmpgt_epi32(qcoeff0, zbin);
cmp_mask1 = _mm_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask, iscan + index, &is_found1, &mask1);
all_zero = _mm_or_si128(cmp_mask0, cmp_mask1);
if (_mm_movemask_epi8(all_zero) == 0) {
_mm_store_si128((__m128i *)(qcoeff_ptr + index), zero);
_mm_store_si128((__m128i *)(qcoeff_ptr + index + 4), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index + 4), zero);
index += 8;
continue;
}
highbd_calculate_qcoeff(&qcoeff0, &round, &quant, &shift, &log_scale);
highbd_calculate_qcoeff(&qcoeff1, &round, &quant, &shift, &log_scale);
qcoeff0 = invert_sign_32_sse2(qcoeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(qcoeff1, coeff1_sign);
qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
_mm_store_si128((__m128i *)(qcoeff_ptr + index), qcoeff0);
_mm_store_si128((__m128i *)(qcoeff_ptr + index + 4), qcoeff1);
coeff0 = highbd_calculate_dqcoeff(qcoeff0, dequant, log_scale);
coeff1 = highbd_calculate_dqcoeff(qcoeff1, dequant, log_scale);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index), coeff0);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index + 4), coeff1);
index += 8;
}
if (is_found0) non_zero_count = calculate_non_zero_count(mask0);
if (is_found1)
non_zero_count_prescan_add_zero = calculate_non_zero_count(mask1);
for (int i = non_zero_count_prescan_add_zero - 1; i >= non_zero_count; i--) {
const int rc = scan[i];
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
}
for (int i = non_zero_count - 1; i >= 0; i--) {
const int rc = scan[i];
if (qcoeff_ptr[rc]) {
eob = i;
break;
}
}
*eob_ptr = eob + 1;
#if SKIP_EOB_FACTOR_ADJUST
// TODO(Aniket): Experiment the following loop with intrinsic by combining
// with the quantization loop above
for (int i = 0; i < non_zero_count; i++) {
const int rc = scan[i];
const int qcoeff = qcoeff_ptr[rc];
if (qcoeff) {
first = i;
break;
}
}
if ((*eob_ptr - 1) >= 0 && first == (*eob_ptr - 1)) {
const int rc = scan[(*eob_ptr - 1)];
if (qcoeff_ptr[rc] == 1 || qcoeff_ptr[rc] == -1) {
const int coeff = coeff_ptr[rc] * wt;
const int coeff_sign = (coeff >> 31);
const int abs_coeff = (coeff ^ coeff_sign) - coeff_sign;
const int factor = EOB_FACTOR + SKIP_EOB_FACTOR_ADJUST;
const int prescan_add_val =
ROUND_POWER_OF_TWO(dequant_ptr[rc != 0] * factor, 7);
if (abs_coeff <
(zbin_ptr[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val)) {
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
*eob_ptr = 0;
}
}
}
#endif
}
void aom_highbd_quantize_b_32x32_adaptive_sse2(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
int index = 8;
const int log_scale = 1;
int non_zero_count = 0;
int non_zero_count_prescan_add_zero = 0;
int is_found0 = 0, is_found1 = 0;
int eob = -1;
const __m128i zero = _mm_setzero_si128();
const __m128i one = _mm_set1_epi32(1);
const __m128i log_scale_vec = _mm_set1_epi32(log_scale);
__m128i zbin, round, quant, dequant, shift;
__m128i coeff0, coeff1, coeff0_sign, coeff1_sign;
__m128i qcoeff0, qcoeff1;
__m128i cmp_mask0, cmp_mask1, cmp_mask;
__m128i all_zero;
__m128i mask0 = zero, mask1 = zero;
const int zbins[2] = { ROUND_POWER_OF_TWO(zbin_ptr[0], log_scale),
ROUND_POWER_OF_TWO(zbin_ptr[1], log_scale) };
int prescan_add[2];
int thresh[4];
const qm_val_t wt = (1 << AOM_QM_BITS);
for (int i = 0; i < 2; ++i) {
prescan_add[i] = ROUND_POWER_OF_TWO(dequant_ptr[i] * EOB_FACTOR, 7);
thresh[i] = (zbins[i] * wt + prescan_add[i]) - 1;
}
thresh[2] = thresh[3] = thresh[1];
__m128i threshold[2];
threshold[0] = _mm_loadu_si128((__m128i *)&thresh[0]);
threshold[1] = _mm_unpackhi_epi64(threshold[0], threshold[0]);
#if SKIP_EOB_FACTOR_ADJUST
int first = -1;
#endif
// Setup global values.
zbin = _mm_load_si128((const __m128i *)zbin_ptr);
round = _mm_load_si128((const __m128i *)round_ptr);
quant = _mm_load_si128((const __m128i *)quant_ptr);
dequant = _mm_load_si128((const __m128i *)dequant_ptr);
shift = _mm_load_si128((const __m128i *)quant_shift_ptr);
__m128i zbin_sign = _mm_srai_epi16(zbin, 15);
__m128i round_sign = _mm_srai_epi16(round, 15);
__m128i quant_sign = _mm_srai_epi16(quant, 15);
__m128i dequant_sign = _mm_srai_epi16(dequant, 15);
__m128i shift_sign = _mm_srai_epi16(shift, 15);
zbin = _mm_unpacklo_epi16(zbin, zbin_sign);
round = _mm_unpacklo_epi16(round, round_sign);
quant = _mm_unpacklo_epi16(quant, quant_sign);
dequant = _mm_unpacklo_epi16(dequant, dequant_sign);
shift = _mm_unpacklo_epi16(shift, shift_sign);
// Shift with rounding.
zbin = _mm_add_epi32(zbin, log_scale_vec);
round = _mm_add_epi32(round, log_scale_vec);
zbin = _mm_srli_epi32(zbin, log_scale);
round = _mm_srli_epi32(round, log_scale);
zbin = _mm_sub_epi32(zbin, one);
// Do DC and first 15 AC.
coeff0 = _mm_load_si128((__m128i *)(coeff_ptr));
coeff1 = _mm_load_si128((__m128i *)(coeff_ptr + 4));
coeff0_sign = _mm_srai_epi32(coeff0, 31);
coeff1_sign = _mm_srai_epi32(coeff1, 31);
qcoeff0 = invert_sign_32_sse2(coeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(coeff1, coeff1_sign);
highbd_update_mask0(&qcoeff0, &qcoeff1, threshold, iscan, &is_found0, &mask0);
cmp_mask0 = _mm_cmpgt_epi32(qcoeff0, zbin);
zbin = _mm_unpackhi_epi64(zbin, zbin); // Switch DC to AC
cmp_mask1 = _mm_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask, iscan, &is_found1, &mask1);
threshold[0] = threshold[1];
all_zero = _mm_or_si128(cmp_mask0, cmp_mask1);
if (_mm_movemask_epi8(all_zero) == 0) {
_mm_store_si128((__m128i *)(qcoeff_ptr), zero);
_mm_store_si128((__m128i *)(qcoeff_ptr + 4), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + 4), zero);
round = _mm_unpackhi_epi64(round, round);
quant = _mm_unpackhi_epi64(quant, quant);
shift = _mm_unpackhi_epi64(shift, shift);
dequant = _mm_unpackhi_epi64(dequant, dequant);
} else {
highbd_calculate_qcoeff(&qcoeff0, &round, &quant, &shift, &log_scale);
round = _mm_unpackhi_epi64(round, round);
quant = _mm_unpackhi_epi64(quant, quant);
shift = _mm_unpackhi_epi64(shift, shift);
highbd_calculate_qcoeff(&qcoeff1, &round, &quant, &shift, &log_scale);
// Reinsert signs
qcoeff0 = invert_sign_32_sse2(qcoeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(qcoeff1, coeff1_sign);
// Mask out zbin threshold coeffs
qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
_mm_store_si128((__m128i *)(qcoeff_ptr), qcoeff0);
_mm_store_si128((__m128i *)(qcoeff_ptr + 4), qcoeff1);
coeff0 = highbd_calculate_dqcoeff(qcoeff0, dequant, log_scale);
dequant = _mm_unpackhi_epi64(dequant, dequant);
coeff1 = highbd_calculate_dqcoeff(qcoeff1, dequant, log_scale);
_mm_store_si128((__m128i *)(dqcoeff_ptr), coeff0);
_mm_store_si128((__m128i *)(dqcoeff_ptr + 4), coeff1);
}
// AC only loop.
while (index < n_coeffs) {
coeff0 = _mm_load_si128((__m128i *)(coeff_ptr + index));
coeff1 = _mm_load_si128((__m128i *)(coeff_ptr + index + 4));
coeff0_sign = _mm_srai_epi32(coeff0, 31);
coeff1_sign = _mm_srai_epi32(coeff1, 31);
qcoeff0 = invert_sign_32_sse2(coeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(coeff1, coeff1_sign);
highbd_update_mask0(&qcoeff0, &qcoeff1, threshold, iscan + index,
&is_found0, &mask0);
cmp_mask0 = _mm_cmpgt_epi32(qcoeff0, zbin);
cmp_mask1 = _mm_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask, iscan + index, &is_found1, &mask1);
all_zero = _mm_or_si128(cmp_mask0, cmp_mask1);
if (_mm_movemask_epi8(all_zero) == 0) {
_mm_store_si128((__m128i *)(qcoeff_ptr + index), zero);
_mm_store_si128((__m128i *)(qcoeff_ptr + index + 4), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index + 4), zero);
index += 8;
continue;
}
highbd_calculate_qcoeff(&qcoeff0, &round, &quant, &shift, &log_scale);
highbd_calculate_qcoeff(&qcoeff1, &round, &quant, &shift, &log_scale);
qcoeff0 = invert_sign_32_sse2(qcoeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(qcoeff1, coeff1_sign);
qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
_mm_store_si128((__m128i *)(qcoeff_ptr + index), qcoeff0);
_mm_store_si128((__m128i *)(qcoeff_ptr + index + 4), qcoeff1);
coeff0 = highbd_calculate_dqcoeff(qcoeff0, dequant, log_scale);
coeff1 = highbd_calculate_dqcoeff(qcoeff1, dequant, log_scale);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index), coeff0);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index + 4), coeff1);
index += 8;
}
if (is_found0) non_zero_count = calculate_non_zero_count(mask0);
if (is_found1)
non_zero_count_prescan_add_zero = calculate_non_zero_count(mask1);
for (int i = non_zero_count_prescan_add_zero - 1; i >= non_zero_count; i--) {
const int rc = scan[i];
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
}
for (int i = non_zero_count - 1; i >= 0; i--) {
const int rc = scan[i];
if (qcoeff_ptr[rc]) {
eob = i;
break;
}
}
*eob_ptr = eob + 1;
#if SKIP_EOB_FACTOR_ADJUST
// TODO(Aniket): Experiment the following loop with intrinsic by combining
// with the quantization loop above
for (int i = 0; i < non_zero_count; i++) {
const int rc = scan[i];
const int qcoeff = qcoeff_ptr[rc];
if (qcoeff) {
first = i;
break;
}
}
if ((*eob_ptr - 1) >= 0 && first == (*eob_ptr - 1)) {
const int rc = scan[(*eob_ptr - 1)];
if (qcoeff_ptr[rc] == 1 || qcoeff_ptr[rc] == -1) {
const int coeff = coeff_ptr[rc] * wt;
const int coeff_sign = (coeff >> 31);
const int abs_coeff = (coeff ^ coeff_sign) - coeff_sign;
const int factor = EOB_FACTOR + SKIP_EOB_FACTOR_ADJUST;
const int prescan_add_val =
ROUND_POWER_OF_TWO(dequant_ptr[rc != 0] * factor, 7);
if (abs_coeff < (zbins[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val)) {
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
*eob_ptr = 0;
}
}
}
#endif
}
void aom_highbd_quantize_b_64x64_adaptive_sse2(
const tran_low_t *coeff_ptr, intptr_t n_coeffs, const int16_t *zbin_ptr,
const int16_t *round_ptr, const int16_t *quant_ptr,
const int16_t *quant_shift_ptr, tran_low_t *qcoeff_ptr,
tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr,
const int16_t *scan, const int16_t *iscan) {
int index = 8;
const int log_scale = 2;
int non_zero_count = 0;
int non_zero_count_prescan_add_zero = 0;
int is_found0 = 0, is_found1 = 0;
int eob = -1;
const __m128i zero = _mm_setzero_si128();
const __m128i one = _mm_set1_epi32(1);
const __m128i log_scale_vec = _mm_set1_epi32(log_scale);
__m128i zbin, round, quant, dequant, shift;
__m128i coeff0, coeff1, coeff0_sign, coeff1_sign;
__m128i qcoeff0, qcoeff1;
__m128i cmp_mask0, cmp_mask1, cmp_mask;
__m128i all_zero;
__m128i mask0 = zero, mask1 = zero;
const int zbins[2] = { ROUND_POWER_OF_TWO(zbin_ptr[0], log_scale),
ROUND_POWER_OF_TWO(zbin_ptr[1], log_scale) };
int prescan_add[2];
int thresh[4];
const qm_val_t wt = (1 << AOM_QM_BITS);
for (int i = 0; i < 2; ++i) {
prescan_add[i] = ROUND_POWER_OF_TWO(dequant_ptr[i] * EOB_FACTOR, 7);
thresh[i] = (zbins[i] * wt + prescan_add[i]) - 1;
}
thresh[2] = thresh[3] = thresh[1];
__m128i threshold[2];
threshold[0] = _mm_loadu_si128((__m128i *)&thresh[0]);
threshold[1] = _mm_unpackhi_epi64(threshold[0], threshold[0]);
#if SKIP_EOB_FACTOR_ADJUST
int first = -1;
#endif
// Setup global values.
zbin = _mm_load_si128((const __m128i *)zbin_ptr);
round = _mm_load_si128((const __m128i *)round_ptr);
quant = _mm_load_si128((const __m128i *)quant_ptr);
dequant = _mm_load_si128((const __m128i *)dequant_ptr);
shift = _mm_load_si128((const __m128i *)quant_shift_ptr);
__m128i zbin_sign = _mm_srai_epi16(zbin, 15);
__m128i round_sign = _mm_srai_epi16(round, 15);
__m128i quant_sign = _mm_srai_epi16(quant, 15);
__m128i dequant_sign = _mm_srai_epi16(dequant, 15);
__m128i shift_sign = _mm_srai_epi16(shift, 15);
zbin = _mm_unpacklo_epi16(zbin, zbin_sign);
round = _mm_unpacklo_epi16(round, round_sign);
quant = _mm_unpacklo_epi16(quant, quant_sign);
dequant = _mm_unpacklo_epi16(dequant, dequant_sign);
shift = _mm_unpacklo_epi16(shift, shift_sign);
// Shift with rounding.
zbin = _mm_add_epi32(zbin, log_scale_vec);
round = _mm_add_epi32(round, log_scale_vec);
zbin = _mm_srli_epi32(zbin, log_scale);
round = _mm_srli_epi32(round, log_scale);
zbin = _mm_sub_epi32(zbin, one);
// Do DC and first 15 AC.
coeff0 = _mm_load_si128((__m128i *)(coeff_ptr));
coeff1 = _mm_load_si128((__m128i *)(coeff_ptr + 4));
coeff0_sign = _mm_srai_epi32(coeff0, 31);
coeff1_sign = _mm_srai_epi32(coeff1, 31);
qcoeff0 = invert_sign_32_sse2(coeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(coeff1, coeff1_sign);
highbd_update_mask0(&qcoeff0, &qcoeff1, threshold, iscan, &is_found0, &mask0);
cmp_mask0 = _mm_cmpgt_epi32(qcoeff0, zbin);
zbin = _mm_unpackhi_epi64(zbin, zbin); // Switch DC to AC
cmp_mask1 = _mm_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask, iscan, &is_found1, &mask1);
threshold[0] = threshold[1];
all_zero = _mm_or_si128(cmp_mask0, cmp_mask1);
if (_mm_movemask_epi8(all_zero) == 0) {
_mm_store_si128((__m128i *)(qcoeff_ptr), zero);
_mm_store_si128((__m128i *)(qcoeff_ptr + 4), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + 4), zero);
round = _mm_unpackhi_epi64(round, round);
quant = _mm_unpackhi_epi64(quant, quant);
shift = _mm_unpackhi_epi64(shift, shift);
dequant = _mm_unpackhi_epi64(dequant, dequant);
} else {
highbd_calculate_qcoeff(&qcoeff0, &round, &quant, &shift, &log_scale);
round = _mm_unpackhi_epi64(round, round);
quant = _mm_unpackhi_epi64(quant, quant);
shift = _mm_unpackhi_epi64(shift, shift);
highbd_calculate_qcoeff(&qcoeff1, &round, &quant, &shift, &log_scale);
// Reinsert signs
qcoeff0 = invert_sign_32_sse2(qcoeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(qcoeff1, coeff1_sign);
// Mask out zbin threshold coeffs
qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
_mm_store_si128((__m128i *)(qcoeff_ptr), qcoeff0);
_mm_store_si128((__m128i *)(qcoeff_ptr + 4), qcoeff1);
coeff0 = highbd_calculate_dqcoeff(qcoeff0, dequant, log_scale);
dequant = _mm_unpackhi_epi64(dequant, dequant);
coeff1 = highbd_calculate_dqcoeff(qcoeff1, dequant, log_scale);
_mm_store_si128((__m128i *)(dqcoeff_ptr), coeff0);
_mm_store_si128((__m128i *)(dqcoeff_ptr + 4), coeff1);
}
// AC only loop.
while (index < n_coeffs) {
coeff0 = _mm_load_si128((__m128i *)(coeff_ptr + index));
coeff1 = _mm_load_si128((__m128i *)(coeff_ptr + index + 4));
coeff0_sign = _mm_srai_epi32(coeff0, 31);
coeff1_sign = _mm_srai_epi32(coeff1, 31);
qcoeff0 = invert_sign_32_sse2(coeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(coeff1, coeff1_sign);
highbd_update_mask0(&qcoeff0, &qcoeff1, threshold, iscan + index,
&is_found0, &mask0);
cmp_mask0 = _mm_cmpgt_epi32(qcoeff0, zbin);
cmp_mask1 = _mm_cmpgt_epi32(qcoeff1, zbin);
cmp_mask = _mm_packs_epi32(cmp_mask0, cmp_mask1);
highbd_update_mask1(&cmp_mask, iscan + index, &is_found1, &mask1);
all_zero = _mm_or_si128(cmp_mask0, cmp_mask1);
if (_mm_movemask_epi8(all_zero) == 0) {
_mm_store_si128((__m128i *)(qcoeff_ptr + index), zero);
_mm_store_si128((__m128i *)(qcoeff_ptr + index + 4), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index), zero);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index + 4), zero);
index += 8;
continue;
}
highbd_calculate_qcoeff(&qcoeff0, &round, &quant, &shift, &log_scale);
highbd_calculate_qcoeff(&qcoeff1, &round, &quant, &shift, &log_scale);
qcoeff0 = invert_sign_32_sse2(qcoeff0, coeff0_sign);
qcoeff1 = invert_sign_32_sse2(qcoeff1, coeff1_sign);
qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
_mm_store_si128((__m128i *)(qcoeff_ptr + index), qcoeff0);
_mm_store_si128((__m128i *)(qcoeff_ptr + index + 4), qcoeff1);
coeff0 = highbd_calculate_dqcoeff(qcoeff0, dequant, log_scale);
coeff1 = highbd_calculate_dqcoeff(qcoeff1, dequant, log_scale);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index), coeff0);
_mm_store_si128((__m128i *)(dqcoeff_ptr + index + 4), coeff1);
index += 8;
}
if (is_found0) non_zero_count = calculate_non_zero_count(mask0);
if (is_found1)
non_zero_count_prescan_add_zero = calculate_non_zero_count(mask1);
for (int i = non_zero_count_prescan_add_zero - 1; i >= non_zero_count; i--) {
const int rc = scan[i];
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
}
for (int i = non_zero_count - 1; i >= 0; i--) {
const int rc = scan[i];
if (qcoeff_ptr[rc]) {
eob = i;
break;
}
}
*eob_ptr = eob + 1;
#if SKIP_EOB_FACTOR_ADJUST
// TODO(Aniket): Experiment the following loop with intrinsic by combining
// with the quantization loop above
for (int i = 0; i < non_zero_count; i++) {
const int rc = scan[i];
const int qcoeff = qcoeff_ptr[rc];
if (qcoeff) {
first = i;
break;
}
}
if ((*eob_ptr - 1) >= 0 && first == (*eob_ptr - 1)) {
const int rc = scan[(*eob_ptr - 1)];
if (qcoeff_ptr[rc] == 1 || qcoeff_ptr[rc] == -1) {
const int coeff = coeff_ptr[rc] * wt;
const int coeff_sign = (coeff >> 31);
const int abs_coeff = (coeff ^ coeff_sign) - coeff_sign;
const int factor = EOB_FACTOR + SKIP_EOB_FACTOR_ADJUST;
const int prescan_add_val =
ROUND_POWER_OF_TWO(dequant_ptr[rc != 0] * factor, 7);
if (abs_coeff < (zbins[rc != 0] * (1 << AOM_QM_BITS) + prescan_add_val)) {
qcoeff_ptr[rc] = 0;
dqcoeff_ptr[rc] = 0;
*eob_ptr = 0;
}
}
}
#endif
}

12
third_party/aom/aom_dsp/x86/highbd_convolve_avx2.c vendored
View file

@ -115,13 +115,13 @@ void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride,
uint16_t *dst, int dst_stride, int w, int h,
const InterpFilterParams *filter_params_x,
const InterpFilterParams *filter_params_y,
const int subpel_x_q4, const int subpel_y_q4,
const int subpel_x_qn, const int subpel_y_qn,
ConvolveParams *conv_params, int bd) {
int i, j;
const int fo_vert = filter_params_y->taps / 2 - 1;
const uint16_t *const src_ptr = src - fo_vert * src_stride;
(void)filter_params_x;
(void)subpel_x_q4;
(void)subpel_x_qn;
(void)conv_params;
assert(conv_params->round_0 <= FILTER_BITS);
@ -138,7 +138,7 @@ void av1_highbd_convolve_y_sr_avx2(const uint16_t *src, int src_stride,
_mm256_set1_epi16(bd == 10 ? 1023 : (bd == 12 ? 4095 : 255));
const __m256i zero = _mm256_setzero_si256();
prepare_coeffs(filter_params_y, subpel_y_q4, coeffs_y);
prepare_coeffs(filter_params_y, subpel_y_qn, coeffs_y);
for (j = 0; j < w; j += 8) {
const uint16_t *data = &src_ptr[j];
@ -264,12 +264,12 @@ void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride,
uint16_t *dst, int dst_stride, int w, int h,
const InterpFilterParams *filter_params_x,
const InterpFilterParams *filter_params_y,
const int subpel_x_q4, const int subpel_y_q4,
const int subpel_x_qn, const int subpel_y_qn,
ConvolveParams *conv_params, int bd) {
int i, j;
const int fo_horiz = filter_params_x->taps / 2 - 1;
const uint16_t *const src_ptr = src - fo_horiz;
(void)subpel_y_q4;
(void)subpel_y_qn;
(void)filter_params_y;
// Check that, even with 12-bit input, the intermediate values will fit
@ -293,7 +293,7 @@ void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride,
assert((FILTER_BITS - conv_params->round_1) >= 0 ||
((conv_params->round_0 + conv_params->round_1) == 2 * FILTER_BITS));
prepare_coeffs(filter_params_x, subpel_x_q4, coeffs_x);
prepare_coeffs(filter_params_x, subpel_x_qn, coeffs_x);
for (j = 0; j < w; j += 8) {
/* Horizontal filter */

16
third_party/aom/aom_dsp/x86/highbd_convolve_ssse3.c vendored
View file

@ -20,14 +20,14 @@ void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride,
uint16_t *dst, int dst_stride, int w, int h,
const InterpFilterParams *filter_params_x,
const InterpFilterParams *filter_params_y,
const int subpel_x_q4,
const int subpel_y_q4,
const int subpel_x_qn,
const int subpel_y_qn,
ConvolveParams *conv_params, int bd) {
int i, j;
const int fo_vert = filter_params_y->taps / 2 - 1;
const uint16_t *const src_ptr = src - fo_vert * src_stride;
(void)filter_params_x;
(void)subpel_x_q4;
(void)subpel_x_qn;
(void)conv_params;
assert(conv_params->round_0 <= FILTER_BITS);
@ -44,7 +44,7 @@ void av1_highbd_convolve_y_sr_ssse3(const uint16_t *src, int src_stride,
_mm_set1_epi16(bd == 10 ? 1023 : (bd == 12 ? 4095 : 255));
const __m128i zero = _mm_setzero_si128();
prepare_coeffs(filter_params_y, subpel_y_q4, coeffs_y);
prepare_coeffs(filter_params_y, subpel_y_qn, coeffs_y);
for (j = 0; j < w; j += 8) {
const uint16_t *data = &src_ptr[j];
@ -168,13 +168,13 @@ void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride,
uint16_t *dst, int dst_stride, int w, int h,
const InterpFilterParams *filter_params_x,
const InterpFilterParams *filter_params_y,
const int subpel_x_q4,
const int subpel_y_q4,
const int subpel_x_qn,
const int subpel_y_qn,
ConvolveParams *conv_params, int bd) {
int i, j;
const int fo_horiz = filter_params_x->taps / 2 - 1;
const uint16_t *const src_ptr = src - fo_horiz;
(void)subpel_y_q4;
(void)subpel_y_qn;
(void)filter_params_y;
// Check that, even with 12-bit input, the intermediate values will fit
@ -195,7 +195,7 @@ void av1_highbd_convolve_x_sr_ssse3(const uint16_t *src, int src_stride,
_mm_set1_epi16(bd == 10 ? 1023 : (bd == 12 ? 4095 : 255));
const __m128i zero = _mm_setzero_si128();
prepare_coeffs(filter_params_x, subpel_x_q4, coeffs_x);
prepare_coeffs(filter_params_x, subpel_x_qn, coeffs_x);
for (j = 0; j < w; j += 8) {
/* Horizontal filter */

Some files were not shown because too many files have changed in this diff Show more