Line data Source code
1 : #ifndef AOM_RTCD_H_
2 : #define AOM_RTCD_H_
3 :
4 : #ifdef RTCD_C
5 : #define RTCD_EXTERN
6 : #else
7 : #define RTCD_EXTERN extern
8 : #endif
9 :
10 : /*
11 : * AV1
12 : */
13 :
14 : #include "aom/aom_integer.h"
15 : #include "av1/common/common.h"
16 : #include "av1/common/enums.h"
17 : #include "av1/common/quant_common.h"
18 : #include "av1/common/filter.h"
19 : #include "av1/common/convolve.h"
20 : #include "av1/common/av1_txfm.h"
21 : #include "av1/common/odintrin.h"
22 :
23 : struct macroblockd;
24 :
25 : /* Encoder forward decls */
26 : struct macroblock;
27 : struct aom_variance_vtable;
28 : struct search_site_config;
29 : struct mv;
30 : union int_mv;
31 : struct yv12_buffer_config;
32 : typedef uint16_t od_dering_in;
33 :
34 : #ifdef __cplusplus
35 : extern "C" {
36 : #endif
37 :
38 : void aom_clpf_block_c(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
39 : void aom_clpf_block_sse2(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
40 : void aom_clpf_block_ssse3(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
41 : void aom_clpf_block_sse4_1(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
42 : RTCD_EXTERN void (*aom_clpf_block)(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
43 :
44 : void aom_clpf_block_hbd_c(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
45 : void aom_clpf_block_hbd_sse2(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
46 : void aom_clpf_block_hbd_ssse3(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
47 : void aom_clpf_block_hbd_sse4_1(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
48 : RTCD_EXTERN void (*aom_clpf_block_hbd)(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
49 :
50 : void aom_clpf_hblock_c(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
51 : void aom_clpf_hblock_sse2(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
52 : void aom_clpf_hblock_ssse3(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
53 : void aom_clpf_hblock_sse4_1(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
54 : RTCD_EXTERN void (*aom_clpf_hblock)(uint8_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
55 :
56 : void aom_clpf_hblock_hbd_c(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
57 : void aom_clpf_hblock_hbd_sse2(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
58 : void aom_clpf_hblock_hbd_ssse3(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
59 : void aom_clpf_hblock_hbd_sse4_1(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
60 : RTCD_EXTERN void (*aom_clpf_hblock_hbd)(uint16_t *dst, const uint16_t *src, int dstride, int sstride, int sizex, int sizey, unsigned int strength, unsigned int bd);
61 :
62 : int64_t av1_block_error_c(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz);
63 : #define av1_block_error av1_block_error_c
64 :
65 : void av1_convolve_horiz_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, ConvolveParams *conv_params);
66 : void av1_convolve_horiz_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, ConvolveParams *conv_params);
67 : RTCD_EXTERN void (*av1_convolve_horiz)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, ConvolveParams *conv_params);
68 :
69 : void av1_convolve_vert_c(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, ConvolveParams *conv_params);
70 : void av1_convolve_vert_ssse3(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, ConvolveParams *conv_params);
71 : RTCD_EXTERN void (*av1_convolve_vert)(const uint8_t *src, int src_stride, uint8_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, ConvolveParams *conv_params);
72 :
73 : int av1_diamond_search_sad_c(struct macroblock *x, const struct search_site_config *cfg, struct mv *ref_mv, struct mv *best_mv, int search_param, int sad_per_bit, int *num00, const struct aom_variance_vtable *fn_ptr, const struct mv *center_mv);
74 : #define av1_diamond_search_sad av1_diamond_search_sad_c
75 :
76 : void av1_fdct8x8_quant_c(const int16_t *input, int stride, tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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);
77 : #define av1_fdct8x8_quant av1_fdct8x8_quant_c
78 :
79 : void av1_fht16x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
80 : void av1_fht16x16_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
81 : void av1_fht16x16_avx2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
82 : RTCD_EXTERN void (*av1_fht16x16)(const int16_t *input, tran_low_t *output, int stride, int tx_type);
83 :
84 : void av1_fht16x32_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
85 : void av1_fht16x32_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
86 : #define av1_fht16x32 av1_fht16x32_sse2
87 :
88 : void av1_fht16x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
89 : #define av1_fht16x4 av1_fht16x4_c
90 :
91 : void av1_fht16x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
92 : void av1_fht16x8_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
93 : #define av1_fht16x8 av1_fht16x8_sse2
94 :
95 : void av1_fht32x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
96 : void av1_fht32x16_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
97 : #define av1_fht32x16 av1_fht32x16_sse2
98 :
99 : void av1_fht32x32_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
100 : void av1_fht32x32_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
101 : void av1_fht32x32_avx2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
102 : RTCD_EXTERN void (*av1_fht32x32)(const int16_t *input, tran_low_t *output, int stride, int tx_type);
103 :
104 : void av1_fht32x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
105 : #define av1_fht32x8 av1_fht32x8_c
106 :
107 : void av1_fht4x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
108 : #define av1_fht4x16 av1_fht4x16_c
109 :
110 : void av1_fht4x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
111 : void av1_fht4x4_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
112 : #define av1_fht4x4 av1_fht4x4_sse2
113 :
114 : void av1_fht4x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
115 : void av1_fht4x8_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
116 : #define av1_fht4x8 av1_fht4x8_sse2
117 :
118 : void av1_fht8x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
119 : void av1_fht8x16_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
120 : #define av1_fht8x16 av1_fht8x16_sse2
121 :
122 : void av1_fht8x32_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
123 : #define av1_fht8x32 av1_fht8x32_c
124 :
125 : void av1_fht8x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
126 : void av1_fht8x4_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
127 : #define av1_fht8x4 av1_fht8x4_sse2
128 :
129 : void av1_fht8x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
130 : void av1_fht8x8_sse2(const int16_t *input, tran_low_t *output, int stride, int tx_type);
131 : #define av1_fht8x8 av1_fht8x8_sse2
132 :
133 : int av1_full_range_search_c(const struct macroblock *x, const struct search_site_config *cfg, struct mv *ref_mv, struct mv *best_mv, int search_param, int sad_per_bit, int *num00, const struct aom_variance_vtable *fn_ptr, const struct mv *center_mv);
134 : #define av1_full_range_search av1_full_range_search_c
135 :
136 : int av1_full_search_sad_c(const struct macroblock *x, const struct mv *ref_mv, int sad_per_bit, int distance, const struct aom_variance_vtable *fn_ptr, const struct mv *center_mv, struct mv *best_mv);
137 : int av1_full_search_sadx3(const struct macroblock *x, const struct mv *ref_mv, int sad_per_bit, int distance, const struct aom_variance_vtable *fn_ptr, const struct mv *center_mv, struct mv *best_mv);
138 : int av1_full_search_sadx8(const struct macroblock *x, const struct mv *ref_mv, int sad_per_bit, int distance, const struct aom_variance_vtable *fn_ptr, const struct mv *center_mv, struct mv *best_mv);
139 : RTCD_EXTERN int (*av1_full_search_sad)(const struct macroblock *x, const struct mv *ref_mv, int sad_per_bit, int distance, const struct aom_variance_vtable *fn_ptr, const struct mv *center_mv, struct mv *best_mv);
140 :
141 : void av1_fwd_idtx_c(const int16_t *src_diff, tran_low_t *coeff, int stride, int bs, int tx_type);
142 : #define av1_fwd_idtx av1_fwd_idtx_c
143 :
144 : void av1_fwd_txfm2d_16x16_c(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
145 : void av1_fwd_txfm2d_16x16_sse4_1(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
146 : RTCD_EXTERN void (*av1_fwd_txfm2d_16x16)(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
147 :
148 : void av1_fwd_txfm2d_32x32_c(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
149 : void av1_fwd_txfm2d_32x32_sse4_1(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
150 : RTCD_EXTERN void (*av1_fwd_txfm2d_32x32)(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
151 :
152 : void av1_fwd_txfm2d_4x4_c(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
153 : void av1_fwd_txfm2d_4x4_sse4_1(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
154 : RTCD_EXTERN void (*av1_fwd_txfm2d_4x4)(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
155 :
156 : void av1_fwd_txfm2d_64x64_c(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
157 : void av1_fwd_txfm2d_64x64_sse4_1(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
158 : RTCD_EXTERN void (*av1_fwd_txfm2d_64x64)(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
159 :
160 : void av1_fwd_txfm2d_8x8_c(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
161 : void av1_fwd_txfm2d_8x8_sse4_1(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
162 : RTCD_EXTERN void (*av1_fwd_txfm2d_8x8)(const int16_t *input, int32_t *output, int stride, int tx_type, int bd);
163 :
164 : void av1_fwht4x4_c(const int16_t *input, tran_low_t *output, int stride);
165 : #define av1_fwht4x4 av1_fwht4x4_c
166 :
167 : int64_t av1_highbd_block_error_c(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz, int bd);
168 : int64_t av1_highbd_block_error_sse2(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz, int bd);
169 : #define av1_highbd_block_error av1_highbd_block_error_sse2
170 :
171 : void av1_highbd_convolve8_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);
172 : void av1_highbd_convolve8_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);
173 : #define av1_highbd_convolve8 av1_highbd_convolve8_sse2
174 :
175 : void av1_highbd_convolve8_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);
176 : void av1_highbd_convolve8_avg_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);
177 : #define av1_highbd_convolve8_avg av1_highbd_convolve8_avg_sse2
178 :
179 : void av1_highbd_convolve8_avg_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);
180 : void av1_highbd_convolve8_avg_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);
181 : #define av1_highbd_convolve8_avg_horiz av1_highbd_convolve8_avg_horiz_sse2
182 :
183 : void av1_highbd_convolve8_avg_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);
184 : void av1_highbd_convolve8_avg_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);
185 : #define av1_highbd_convolve8_avg_vert av1_highbd_convolve8_avg_vert_sse2
186 :
187 : void av1_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);
188 : void av1_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);
189 : #define av1_highbd_convolve8_horiz av1_highbd_convolve8_horiz_sse2
190 :
191 : 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);
192 : 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);
193 : #define av1_highbd_convolve8_vert av1_highbd_convolve8_vert_sse2
194 :
195 : 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);
196 : #define av1_highbd_convolve_avg av1_highbd_convolve_avg_c
197 :
198 : void av1_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);
199 : #define av1_highbd_convolve_copy av1_highbd_convolve_copy_c
200 :
201 : void av1_highbd_convolve_horiz_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, int avg, int bd);
202 : void av1_highbd_convolve_horiz_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, int avg, int bd);
203 : RTCD_EXTERN void (*av1_highbd_convolve_horiz)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, int avg, int bd);
204 :
205 : void av1_highbd_convolve_init_c(void);
206 : void av1_highbd_convolve_init_sse4_1(void);
207 : RTCD_EXTERN void (*av1_highbd_convolve_init)(void);
208 :
209 : void av1_highbd_convolve_vert_c(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, int avg, int bd);
210 : void av1_highbd_convolve_vert_sse4_1(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, int avg, int bd);
211 : RTCD_EXTERN void (*av1_highbd_convolve_vert)(const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, const InterpFilterParams fp, const int subpel_x_q4, int x_step_q4, int avg, int bd);
212 :
213 : void av1_highbd_fht16x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
214 : #define av1_highbd_fht16x16 av1_highbd_fht16x16_c
215 :
216 : void av1_highbd_fht16x32_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
217 : #define av1_highbd_fht16x32 av1_highbd_fht16x32_c
218 :
219 : void av1_highbd_fht16x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
220 : #define av1_highbd_fht16x4 av1_highbd_fht16x4_c
221 :
222 : void av1_highbd_fht16x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
223 : #define av1_highbd_fht16x8 av1_highbd_fht16x8_c
224 :
225 : void av1_highbd_fht32x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
226 : #define av1_highbd_fht32x16 av1_highbd_fht32x16_c
227 :
228 : void av1_highbd_fht32x32_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
229 : #define av1_highbd_fht32x32 av1_highbd_fht32x32_c
230 :
231 : void av1_highbd_fht32x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
232 : #define av1_highbd_fht32x8 av1_highbd_fht32x8_c
233 :
234 : void av1_highbd_fht4x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
235 : #define av1_highbd_fht4x16 av1_highbd_fht4x16_c
236 :
237 : void av1_highbd_fht4x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
238 : void av1_highbd_fht4x4_sse4_1(const int16_t *input, tran_low_t *output, int stride, int tx_type);
239 : RTCD_EXTERN void (*av1_highbd_fht4x4)(const int16_t *input, tran_low_t *output, int stride, int tx_type);
240 :
241 : void av1_highbd_fht4x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
242 : #define av1_highbd_fht4x8 av1_highbd_fht4x8_c
243 :
244 : void av1_highbd_fht8x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
245 : #define av1_highbd_fht8x16 av1_highbd_fht8x16_c
246 :
247 : void av1_highbd_fht8x32_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
248 : #define av1_highbd_fht8x32 av1_highbd_fht8x32_c
249 :
250 : void av1_highbd_fht8x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
251 : #define av1_highbd_fht8x4 av1_highbd_fht8x4_c
252 :
253 : void av1_highbd_fht8x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type);
254 : #define av1_highbd_fht8x8 av1_highbd_fht8x8_c
255 :
256 : void av1_highbd_fwht4x4_c(const int16_t *input, tran_low_t *output, int stride);
257 : #define av1_highbd_fwht4x4 av1_highbd_fwht4x4_c
258 :
259 : void av1_highbd_iht16x16_256_add_c(const tran_low_t *input, uint8_t *output, int pitch, int tx_type, int bd);
260 : #define av1_highbd_iht16x16_256_add av1_highbd_iht16x16_256_add_c
261 :
262 : void av1_highbd_iht16x32_512_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
263 : #define av1_highbd_iht16x32_512_add av1_highbd_iht16x32_512_add_c
264 :
265 : void av1_highbd_iht16x4_64_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
266 : #define av1_highbd_iht16x4_64_add av1_highbd_iht16x4_64_add_c
267 :
268 : void av1_highbd_iht16x8_128_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
269 : #define av1_highbd_iht16x8_128_add av1_highbd_iht16x8_128_add_c
270 :
271 : void av1_highbd_iht32x16_512_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
272 : #define av1_highbd_iht32x16_512_add av1_highbd_iht32x16_512_add_c
273 :
274 : void av1_highbd_iht32x8_256_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
275 : #define av1_highbd_iht32x8_256_add av1_highbd_iht32x8_256_add_c
276 :
277 : void av1_highbd_iht4x16_64_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
278 : #define av1_highbd_iht4x16_64_add av1_highbd_iht4x16_64_add_c
279 :
280 : void av1_highbd_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
281 : #define av1_highbd_iht4x4_16_add av1_highbd_iht4x4_16_add_c
282 :
283 : void av1_highbd_iht4x8_32_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
284 : #define av1_highbd_iht4x8_32_add av1_highbd_iht4x8_32_add_c
285 :
286 : void av1_highbd_iht8x16_128_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
287 : #define av1_highbd_iht8x16_128_add av1_highbd_iht8x16_128_add_c
288 :
289 : void av1_highbd_iht8x32_256_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
290 : #define av1_highbd_iht8x32_256_add av1_highbd_iht8x32_256_add_c
291 :
292 : void av1_highbd_iht8x4_32_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
293 : #define av1_highbd_iht8x4_32_add av1_highbd_iht8x4_32_add_c
294 :
295 : void av1_highbd_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type, int bd);
296 : #define av1_highbd_iht8x8_64_add av1_highbd_iht8x8_64_add_c
297 :
298 : void av1_highbd_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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 log_scale);
299 : #define av1_highbd_quantize_b av1_highbd_quantize_b_c
300 :
301 : void av1_highbd_quantize_fp_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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 log_scale);
302 : void av1_highbd_quantize_fp_sse4_1(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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 log_scale);
303 : RTCD_EXTERN void (*av1_highbd_quantize_fp)(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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 log_scale);
304 :
305 : void av1_highbd_temporal_filter_apply_c(uint8_t *frame1, unsigned int stride, uint8_t *frame2, unsigned int block_width, unsigned int block_height, int strength, int filter_weight, unsigned int *accumulator, uint16_t *count);
306 : #define av1_highbd_temporal_filter_apply av1_highbd_temporal_filter_apply_c
307 :
308 : 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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
309 : void av1_highbd_warp_affine_ssse3(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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
310 : 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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
311 :
312 : void av1_iht16x16_256_add_c(const tran_low_t *input, uint8_t *output, int pitch, int tx_type);
313 : void av1_iht16x16_256_add_sse2(const tran_low_t *input, uint8_t *output, int pitch, int tx_type);
314 : void av1_iht16x16_256_add_avx2(const tran_low_t *input, uint8_t *output, int pitch, int tx_type);
315 : RTCD_EXTERN void (*av1_iht16x16_256_add)(const tran_low_t *input, uint8_t *output, int pitch, int tx_type);
316 :
317 : void av1_iht16x32_512_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
318 : void av1_iht16x32_512_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
319 : #define av1_iht16x32_512_add av1_iht16x32_512_add_sse2
320 :
321 : void av1_iht16x4_64_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
322 : #define av1_iht16x4_64_add av1_iht16x4_64_add_c
323 :
324 : void av1_iht16x8_128_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
325 : void av1_iht16x8_128_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
326 : #define av1_iht16x8_128_add av1_iht16x8_128_add_sse2
327 :
328 : void av1_iht32x16_512_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
329 : void av1_iht32x16_512_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
330 : #define av1_iht32x16_512_add av1_iht32x16_512_add_sse2
331 :
332 : void av1_iht32x32_1024_add_c(const tran_low_t *input, uint8_t *output, int pitch, int tx_type);
333 : #define av1_iht32x32_1024_add av1_iht32x32_1024_add_c
334 :
335 : void av1_iht32x8_256_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
336 : #define av1_iht32x8_256_add av1_iht32x8_256_add_c
337 :
338 : void av1_iht4x16_64_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
339 : #define av1_iht4x16_64_add av1_iht4x16_64_add_c
340 :
341 : void av1_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
342 : void av1_iht4x4_16_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
343 : #define av1_iht4x4_16_add av1_iht4x4_16_add_sse2
344 :
345 : void av1_iht4x8_32_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
346 : void av1_iht4x8_32_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
347 : #define av1_iht4x8_32_add av1_iht4x8_32_add_sse2
348 :
349 : void av1_iht8x16_128_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
350 : void av1_iht8x16_128_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
351 : #define av1_iht8x16_128_add av1_iht8x16_128_add_sse2
352 :
353 : void av1_iht8x32_256_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
354 : #define av1_iht8x32_256_add av1_iht8x32_256_add_c
355 :
356 : void av1_iht8x4_32_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
357 : void av1_iht8x4_32_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
358 : #define av1_iht8x4_32_add av1_iht8x4_32_add_sse2
359 :
360 : void av1_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
361 : void av1_iht8x8_64_add_sse2(const tran_low_t *input, uint8_t *dest, int dest_stride, int tx_type);
362 : #define av1_iht8x8_64_add av1_iht8x8_64_add_sse2
363 :
364 : void av1_inv_txfm2d_add_16x16_c(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
365 : void av1_inv_txfm2d_add_16x16_sse4_1(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
366 : RTCD_EXTERN void (*av1_inv_txfm2d_add_16x16)(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
367 :
368 : void av1_inv_txfm2d_add_32x32_c(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
369 : void av1_inv_txfm2d_add_32x32_avx2(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
370 : RTCD_EXTERN void (*av1_inv_txfm2d_add_32x32)(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
371 :
372 : void av1_inv_txfm2d_add_4x4_c(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
373 : void av1_inv_txfm2d_add_4x4_sse4_1(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
374 : RTCD_EXTERN void (*av1_inv_txfm2d_add_4x4)(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
375 :
376 : void av1_inv_txfm2d_add_64x64_c(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
377 : #define av1_inv_txfm2d_add_64x64 av1_inv_txfm2d_add_64x64_c
378 :
379 : void av1_inv_txfm2d_add_8x8_c(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
380 : void av1_inv_txfm2d_add_8x8_sse4_1(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
381 : RTCD_EXTERN void (*av1_inv_txfm2d_add_8x8)(const int32_t *input, uint16_t *output, int stride, int tx_type, int bd);
382 :
383 : void av1_lowbd_convolve_init_c(void);
384 : void av1_lowbd_convolve_init_ssse3(void);
385 : RTCD_EXTERN void (*av1_lowbd_convolve_init)(void);
386 :
387 : void av1_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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 log_scale);
388 : #define av1_quantize_b av1_quantize_b_c
389 :
390 : void av1_quantize_fp_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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);
391 : void av1_quantize_fp_sse2(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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);
392 : #define av1_quantize_fp av1_quantize_fp_sse2
393 :
394 : void av1_quantize_fp_32x32_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, int skip_block, 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);
395 : #define av1_quantize_fp_32x32 av1_quantize_fp_32x32_c
396 :
397 : void av1_temporal_filter_apply_c(uint8_t *frame1, unsigned int stride, uint8_t *frame2, unsigned int block_width, unsigned int block_height, int strength, int filter_weight, unsigned int *accumulator, uint16_t *count);
398 : void av1_temporal_filter_apply_sse2(uint8_t *frame1, unsigned int stride, uint8_t *frame2, unsigned int block_width, unsigned int block_height, int strength, int filter_weight, unsigned int *accumulator, uint16_t *count);
399 : #define av1_temporal_filter_apply av1_temporal_filter_apply_sse2
400 :
401 : 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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
402 : void av1_warp_affine_sse2(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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
403 : void av1_warp_affine_ssse3(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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
404 : 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, int comp_avg, int16_t alpha, int16_t beta, int16_t gamma, int16_t delta);
405 :
406 : void av1_wedge_compute_delta_squares_c(int16_t *d, const int16_t *a, const int16_t *b, int N);
407 : void av1_wedge_compute_delta_squares_sse2(int16_t *d, const int16_t *a, const int16_t *b, int N);
408 : #define av1_wedge_compute_delta_squares av1_wedge_compute_delta_squares_sse2
409 :
410 : int av1_wedge_sign_from_residuals_c(const int16_t *ds, const uint8_t *m, int N, int64_t limit);
411 : int av1_wedge_sign_from_residuals_sse2(const int16_t *ds, const uint8_t *m, int N, int64_t limit);
412 : #define av1_wedge_sign_from_residuals av1_wedge_sign_from_residuals_sse2
413 :
414 : uint64_t av1_wedge_sse_from_residuals_c(const int16_t *r1, const int16_t *d, const uint8_t *m, int N);
415 : uint64_t av1_wedge_sse_from_residuals_sse2(const int16_t *r1, const int16_t *d, const uint8_t *m, int N);
416 : #define av1_wedge_sse_from_residuals av1_wedge_sse_from_residuals_sse2
417 :
418 : double compute_cross_correlation_c(unsigned char *im1, int stride1, int x1, int y1, unsigned char *im2, int stride2, int x2, int y2);
419 : double compute_cross_correlation_sse4_1(unsigned char *im1, int stride1, int x1, int y1, unsigned char *im2, int stride2, int x2, int y2);
420 : RTCD_EXTERN double (*compute_cross_correlation)(unsigned char *im1, int stride1, int x1, int y1, unsigned char *im2, int stride2, int x2, int y2);
421 :
422 : void copy_4x4_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
423 : void copy_4x4_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
424 : void copy_4x4_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
425 : void copy_4x4_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
426 : RTCD_EXTERN void (*copy_4x4_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
427 :
428 : void copy_4x4_16bit_to_8bit_c(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
429 : void copy_4x4_16bit_to_8bit_sse2(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
430 : void copy_4x4_16bit_to_8bit_ssse3(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
431 : void copy_4x4_16bit_to_8bit_sse4_1(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
432 : RTCD_EXTERN void (*copy_4x4_16bit_to_8bit)(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
433 :
434 : void copy_8x8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
435 : void copy_8x8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
436 : void copy_8x8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
437 : void copy_8x8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
438 : RTCD_EXTERN void (*copy_8x8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride);
439 :
440 : void copy_8x8_16bit_to_8bit_c(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
441 : void copy_8x8_16bit_to_8bit_sse2(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
442 : void copy_8x8_16bit_to_8bit_ssse3(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
443 : void copy_8x8_16bit_to_8bit_sse4_1(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
444 : RTCD_EXTERN void (*copy_8x8_16bit_to_8bit)(uint8_t *dst, int dstride, const uint16_t *src, int sstride);
445 :
446 : void copy_rect8_16bit_to_16bit_c(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
447 : void copy_rect8_16bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
448 : void copy_rect8_16bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
449 : void copy_rect8_16bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
450 : RTCD_EXTERN void (*copy_rect8_16bit_to_16bit)(uint16_t *dst, int dstride, const uint16_t *src, int sstride, int v, int h);
451 :
452 : void copy_rect8_8bit_to_16bit_c(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
453 : void copy_rect8_8bit_to_16bit_sse2(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
454 : void copy_rect8_8bit_to_16bit_ssse3(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
455 : void copy_rect8_8bit_to_16bit_sse4_1(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
456 : RTCD_EXTERN void (*copy_rect8_8bit_to_16bit)(uint16_t *dst, int dstride, const uint8_t *src, int sstride, int v, int h);
457 :
458 : int od_dir_find8_c(const od_dering_in *img, int stride, int32_t *var, int coeff_shift);
459 : int od_dir_find8_sse2(const od_dering_in *img, int stride, int32_t *var, int coeff_shift);
460 : int od_dir_find8_ssse3(const od_dering_in *img, int stride, int32_t *var, int coeff_shift);
461 : int od_dir_find8_sse4_1(const od_dering_in *img, int stride, int32_t *var, int coeff_shift);
462 : RTCD_EXTERN int (*od_dir_find8)(const od_dering_in *img, int stride, int32_t *var, int coeff_shift);
463 :
464 : void od_filter_dering_direction_4x4_c(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
465 : void od_filter_dering_direction_4x4_sse2(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
466 : void od_filter_dering_direction_4x4_ssse3(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
467 : void od_filter_dering_direction_4x4_sse4_1(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
468 : RTCD_EXTERN void (*od_filter_dering_direction_4x4)(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
469 :
470 : void od_filter_dering_direction_8x8_c(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
471 : void od_filter_dering_direction_8x8_sse2(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
472 : void od_filter_dering_direction_8x8_ssse3(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
473 : void od_filter_dering_direction_8x8_sse4_1(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
474 : RTCD_EXTERN void (*od_filter_dering_direction_8x8)(uint16_t *y, int ystride, const uint16_t *in, int threshold, int dir, int damping);
475 :
476 : void aom_rtcd(void);
477 :
478 : #ifdef RTCD_C
479 : #include "aom_ports/x86.h"
480 0 : static void setup_rtcd_internal(void)
481 : {
482 0 : int flags = x86_simd_caps();
483 :
484 : (void)flags;
485 :
486 0 : aom_clpf_block = aom_clpf_block_sse2;
487 0 : if (flags & HAS_SSSE3) aom_clpf_block = aom_clpf_block_ssse3;
488 0 : if (flags & HAS_SSE4_1) aom_clpf_block = aom_clpf_block_sse4_1;
489 0 : aom_clpf_block_hbd = aom_clpf_block_hbd_sse2;
490 0 : if (flags & HAS_SSSE3) aom_clpf_block_hbd = aom_clpf_block_hbd_ssse3;
491 0 : if (flags & HAS_SSE4_1) aom_clpf_block_hbd = aom_clpf_block_hbd_sse4_1;
492 0 : aom_clpf_hblock = aom_clpf_hblock_sse2;
493 0 : if (flags & HAS_SSSE3) aom_clpf_hblock = aom_clpf_hblock_ssse3;
494 0 : if (flags & HAS_SSE4_1) aom_clpf_hblock = aom_clpf_hblock_sse4_1;
495 0 : aom_clpf_hblock_hbd = aom_clpf_hblock_hbd_sse2;
496 0 : if (flags & HAS_SSSE3) aom_clpf_hblock_hbd = aom_clpf_hblock_hbd_ssse3;
497 0 : if (flags & HAS_SSE4_1) aom_clpf_hblock_hbd = aom_clpf_hblock_hbd_sse4_1;
498 0 : av1_convolve_horiz = av1_convolve_horiz_c;
499 0 : if (flags & HAS_SSSE3) av1_convolve_horiz = av1_convolve_horiz_ssse3;
500 0 : av1_convolve_vert = av1_convolve_vert_c;
501 0 : if (flags & HAS_SSSE3) av1_convolve_vert = av1_convolve_vert_ssse3;
502 0 : av1_fht16x16 = av1_fht16x16_sse2;
503 0 : if (flags & HAS_AVX2) av1_fht16x16 = av1_fht16x16_avx2;
504 0 : av1_fht32x32 = av1_fht32x32_sse2;
505 0 : if (flags & HAS_AVX2) av1_fht32x32 = av1_fht32x32_avx2;
506 0 : av1_full_search_sad = av1_full_search_sad_c;
507 0 : if (flags & HAS_SSE3) av1_full_search_sad = av1_full_search_sadx3;
508 0 : if (flags & HAS_SSE4_1) av1_full_search_sad = av1_full_search_sadx8;
509 0 : av1_fwd_txfm2d_16x16 = av1_fwd_txfm2d_16x16_c;
510 0 : if (flags & HAS_SSE4_1) av1_fwd_txfm2d_16x16 = av1_fwd_txfm2d_16x16_sse4_1;
511 0 : av1_fwd_txfm2d_32x32 = av1_fwd_txfm2d_32x32_c;
512 0 : if (flags & HAS_SSE4_1) av1_fwd_txfm2d_32x32 = av1_fwd_txfm2d_32x32_sse4_1;
513 0 : av1_fwd_txfm2d_4x4 = av1_fwd_txfm2d_4x4_c;
514 0 : if (flags & HAS_SSE4_1) av1_fwd_txfm2d_4x4 = av1_fwd_txfm2d_4x4_sse4_1;
515 0 : av1_fwd_txfm2d_64x64 = av1_fwd_txfm2d_64x64_c;
516 0 : if (flags & HAS_SSE4_1) av1_fwd_txfm2d_64x64 = av1_fwd_txfm2d_64x64_sse4_1;
517 0 : av1_fwd_txfm2d_8x8 = av1_fwd_txfm2d_8x8_c;
518 0 : if (flags & HAS_SSE4_1) av1_fwd_txfm2d_8x8 = av1_fwd_txfm2d_8x8_sse4_1;
519 0 : av1_highbd_convolve_horiz = av1_highbd_convolve_horiz_c;
520 0 : if (flags & HAS_SSE4_1) av1_highbd_convolve_horiz = av1_highbd_convolve_horiz_sse4_1;
521 0 : av1_highbd_convolve_init = av1_highbd_convolve_init_c;
522 0 : if (flags & HAS_SSE4_1) av1_highbd_convolve_init = av1_highbd_convolve_init_sse4_1;
523 0 : av1_highbd_convolve_vert = av1_highbd_convolve_vert_c;
524 0 : if (flags & HAS_SSE4_1) av1_highbd_convolve_vert = av1_highbd_convolve_vert_sse4_1;
525 0 : av1_highbd_fht4x4 = av1_highbd_fht4x4_c;
526 0 : if (flags & HAS_SSE4_1) av1_highbd_fht4x4 = av1_highbd_fht4x4_sse4_1;
527 0 : av1_highbd_quantize_fp = av1_highbd_quantize_fp_c;
528 0 : if (flags & HAS_SSE4_1) av1_highbd_quantize_fp = av1_highbd_quantize_fp_sse4_1;
529 0 : av1_highbd_warp_affine = av1_highbd_warp_affine_c;
530 0 : if (flags & HAS_SSSE3) av1_highbd_warp_affine = av1_highbd_warp_affine_ssse3;
531 0 : av1_iht16x16_256_add = av1_iht16x16_256_add_sse2;
532 0 : if (flags & HAS_AVX2) av1_iht16x16_256_add = av1_iht16x16_256_add_avx2;
533 0 : av1_inv_txfm2d_add_16x16 = av1_inv_txfm2d_add_16x16_c;
534 0 : if (flags & HAS_SSE4_1) av1_inv_txfm2d_add_16x16 = av1_inv_txfm2d_add_16x16_sse4_1;
535 0 : av1_inv_txfm2d_add_32x32 = av1_inv_txfm2d_add_32x32_c;
536 0 : if (flags & HAS_AVX2) av1_inv_txfm2d_add_32x32 = av1_inv_txfm2d_add_32x32_avx2;
537 0 : av1_inv_txfm2d_add_4x4 = av1_inv_txfm2d_add_4x4_c;
538 0 : if (flags & HAS_SSE4_1) av1_inv_txfm2d_add_4x4 = av1_inv_txfm2d_add_4x4_sse4_1;
539 0 : av1_inv_txfm2d_add_8x8 = av1_inv_txfm2d_add_8x8_c;
540 0 : if (flags & HAS_SSE4_1) av1_inv_txfm2d_add_8x8 = av1_inv_txfm2d_add_8x8_sse4_1;
541 0 : av1_lowbd_convolve_init = av1_lowbd_convolve_init_c;
542 0 : if (flags & HAS_SSSE3) av1_lowbd_convolve_init = av1_lowbd_convolve_init_ssse3;
543 0 : av1_warp_affine = av1_warp_affine_sse2;
544 0 : if (flags & HAS_SSSE3) av1_warp_affine = av1_warp_affine_ssse3;
545 0 : compute_cross_correlation = compute_cross_correlation_c;
546 0 : if (flags & HAS_SSE4_1) compute_cross_correlation = compute_cross_correlation_sse4_1;
547 0 : copy_4x4_16bit_to_16bit = copy_4x4_16bit_to_16bit_sse2;
548 0 : if (flags & HAS_SSSE3) copy_4x4_16bit_to_16bit = copy_4x4_16bit_to_16bit_ssse3;
549 0 : if (flags & HAS_SSE4_1) copy_4x4_16bit_to_16bit = copy_4x4_16bit_to_16bit_sse4_1;
550 0 : copy_4x4_16bit_to_8bit = copy_4x4_16bit_to_8bit_sse2;
551 0 : if (flags & HAS_SSSE3) copy_4x4_16bit_to_8bit = copy_4x4_16bit_to_8bit_ssse3;
552 0 : if (flags & HAS_SSE4_1) copy_4x4_16bit_to_8bit = copy_4x4_16bit_to_8bit_sse4_1;
553 0 : copy_8x8_16bit_to_16bit = copy_8x8_16bit_to_16bit_sse2;
554 0 : if (flags & HAS_SSSE3) copy_8x8_16bit_to_16bit = copy_8x8_16bit_to_16bit_ssse3;
555 0 : if (flags & HAS_SSE4_1) copy_8x8_16bit_to_16bit = copy_8x8_16bit_to_16bit_sse4_1;
556 0 : copy_8x8_16bit_to_8bit = copy_8x8_16bit_to_8bit_sse2;
557 0 : if (flags & HAS_SSSE3) copy_8x8_16bit_to_8bit = copy_8x8_16bit_to_8bit_ssse3;
558 0 : if (flags & HAS_SSE4_1) copy_8x8_16bit_to_8bit = copy_8x8_16bit_to_8bit_sse4_1;
559 0 : copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse2;
560 0 : if (flags & HAS_SSSE3) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_ssse3;
561 0 : if (flags & HAS_SSE4_1) copy_rect8_16bit_to_16bit = copy_rect8_16bit_to_16bit_sse4_1;
562 0 : copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse2;
563 0 : if (flags & HAS_SSSE3) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_ssse3;
564 0 : if (flags & HAS_SSE4_1) copy_rect8_8bit_to_16bit = copy_rect8_8bit_to_16bit_sse4_1;
565 0 : od_dir_find8 = od_dir_find8_sse2;
566 0 : if (flags & HAS_SSSE3) od_dir_find8 = od_dir_find8_ssse3;
567 0 : if (flags & HAS_SSE4_1) od_dir_find8 = od_dir_find8_sse4_1;
568 0 : od_filter_dering_direction_4x4 = od_filter_dering_direction_4x4_sse2;
569 0 : if (flags & HAS_SSSE3) od_filter_dering_direction_4x4 = od_filter_dering_direction_4x4_ssse3;
570 0 : if (flags & HAS_SSE4_1) od_filter_dering_direction_4x4 = od_filter_dering_direction_4x4_sse4_1;
571 0 : od_filter_dering_direction_8x8 = od_filter_dering_direction_8x8_sse2;
572 0 : if (flags & HAS_SSSE3) od_filter_dering_direction_8x8 = od_filter_dering_direction_8x8_ssse3;
573 0 : if (flags & HAS_SSE4_1) od_filter_dering_direction_8x8 = od_filter_dering_direction_8x8_sse4_1;
574 0 : }
575 : #endif
576 :
577 : #ifdef __cplusplus
578 : } // extern "C"
579 : #endif
580 :
581 : #endif
|
