Mercurial > libavcodec.hg
annotate arm/dsputil_init_neon.c @ 11557:53822d92c3f7 libavcodec
Make sure the EC code does not attempt to use inter based concealment if there
is no reference frame available. (this can happen because the EC code will attempt
to use reference frames even for I/IDR frames)
author | michael |
---|---|
date | Tue, 30 Mar 2010 20:46:46 +0000 |
parents | 2a4dc3c0b012 |
children | f7281af560fe |
rev | line source |
---|---|
8334 | 1 /* |
2 * ARM NEON optimised DSP functions | |
3 * Copyright (c) 2008 Mans Rullgard <mans@mansr.com> | |
4 * | |
5 * This file is part of FFmpeg. | |
6 * | |
7 * FFmpeg is free software; you can redistribute it and/or | |
8 * modify it under the terms of the GNU Lesser General Public | |
9 * License as published by the Free Software Foundation; either | |
10 * version 2.1 of the License, or (at your option) any later version. | |
11 * | |
12 * FFmpeg is distributed in the hope that it will be useful, | |
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
15 * Lesser General Public License for more details. | |
16 * | |
17 * You should have received a copy of the GNU Lesser General Public | |
18 * License along with FFmpeg; if not, write to the Free Software | |
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | |
20 */ | |
21 | |
22 #include <stdint.h> | |
23 | |
24 #include "libavcodec/avcodec.h" | |
25 #include "libavcodec/dsputil.h" | |
10359 | 26 #include "dsputil_arm.h" |
27 | |
28 void ff_simple_idct_neon(DCTELEM *data); | |
29 void ff_simple_idct_put_neon(uint8_t *dest, int line_size, DCTELEM *data); | |
30 void ff_simple_idct_add_neon(uint8_t *dest, int line_size, DCTELEM *data); | |
31 | |
32 void ff_vp3_idct_neon(DCTELEM *data); | |
33 void ff_vp3_idct_put_neon(uint8_t *dest, int line_size, DCTELEM *data); | |
34 void ff_vp3_idct_add_neon(uint8_t *dest, int line_size, DCTELEM *data); | |
8334 | 35 |
36 void ff_put_pixels16_neon(uint8_t *, const uint8_t *, int, int); | |
37 void ff_put_pixels16_x2_neon(uint8_t *, const uint8_t *, int, int); | |
38 void ff_put_pixels16_y2_neon(uint8_t *, const uint8_t *, int, int); | |
39 void ff_put_pixels16_xy2_neon(uint8_t *, const uint8_t *, int, int); | |
40 void ff_put_pixels8_neon(uint8_t *, const uint8_t *, int, int); | |
41 void ff_put_pixels8_x2_neon(uint8_t *, const uint8_t *, int, int); | |
42 void ff_put_pixels8_y2_neon(uint8_t *, const uint8_t *, int, int); | |
43 void ff_put_pixels8_xy2_neon(uint8_t *, const uint8_t *, int, int); | |
44 void ff_put_pixels16_x2_no_rnd_neon(uint8_t *, const uint8_t *, int, int); | |
45 void ff_put_pixels16_y2_no_rnd_neon(uint8_t *, const uint8_t *, int, int); | |
46 void ff_put_pixels16_xy2_no_rnd_neon(uint8_t *, const uint8_t *, int, int); | |
47 void ff_put_pixels8_x2_no_rnd_neon(uint8_t *, const uint8_t *, int, int); | |
48 void ff_put_pixels8_y2_no_rnd_neon(uint8_t *, const uint8_t *, int, int); | |
49 void ff_put_pixels8_xy2_no_rnd_neon(uint8_t *, const uint8_t *, int, int); | |
50 | |
51 void ff_avg_pixels16_neon(uint8_t *, const uint8_t *, int, int); | |
10375 | 52 void ff_avg_pixels8_neon(uint8_t *, const uint8_t *, int, int); |
8334 | 53 |
9344 | 54 void ff_add_pixels_clamped_neon(const DCTELEM *, uint8_t *, int); |
9580 | 55 void ff_put_pixels_clamped_neon(const DCTELEM *, uint8_t *, int); |
9345 | 56 void ff_put_signed_pixels_clamped_neon(const DCTELEM *, uint8_t *, int); |
9344 | 57 |
8334 | 58 void ff_put_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int); |
8338 | 59 void ff_put_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int); |
60 void ff_put_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int); | |
61 void ff_put_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int); | |
62 void ff_put_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int); | |
63 void ff_put_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int); | |
64 void ff_put_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int); | |
65 void ff_put_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int); | |
66 void ff_put_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int); | |
67 void ff_put_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int); | |
68 void ff_put_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int); | |
69 void ff_put_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int); | |
70 void ff_put_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int); | |
71 void ff_put_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int); | |
72 void ff_put_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int); | |
73 void ff_put_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int); | |
74 | |
8334 | 75 void ff_put_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int); |
8338 | 76 void ff_put_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int); |
77 void ff_put_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int); | |
78 void ff_put_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int); | |
79 void ff_put_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int); | |
80 void ff_put_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int); | |
81 void ff_put_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int); | |
82 void ff_put_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int); | |
83 void ff_put_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int); | |
84 void ff_put_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int); | |
85 void ff_put_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int); | |
86 void ff_put_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int); | |
87 void ff_put_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int); | |
88 void ff_put_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int); | |
89 void ff_put_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int); | |
90 void ff_put_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int); | |
8334 | 91 |
92 void ff_avg_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int); | |
10616 | 93 void ff_avg_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int); |
94 void ff_avg_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int); | |
95 void ff_avg_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int); | |
96 void ff_avg_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int); | |
97 void ff_avg_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int); | |
98 void ff_avg_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int); | |
99 void ff_avg_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int); | |
100 void ff_avg_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int); | |
101 void ff_avg_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int); | |
102 void ff_avg_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int); | |
103 void ff_avg_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int); | |
104 void ff_avg_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int); | |
105 void ff_avg_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int); | |
106 void ff_avg_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int); | |
107 void ff_avg_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int); | |
8334 | 108 |
10375 | 109 void ff_avg_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int); |
10616 | 110 void ff_avg_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int); |
111 void ff_avg_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int); | |
112 void ff_avg_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int); | |
113 void ff_avg_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int); | |
114 void ff_avg_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int); | |
115 void ff_avg_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int); | |
116 void ff_avg_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int); | |
117 void ff_avg_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int); | |
118 void ff_avg_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int); | |
119 void ff_avg_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int); | |
120 void ff_avg_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int); | |
121 void ff_avg_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int); | |
122 void ff_avg_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int); | |
123 void ff_avg_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int); | |
124 void ff_avg_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int); | |
10375 | 125 |
8336 | 126 void ff_put_h264_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int); |
127 void ff_put_h264_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int); | |
10617 | 128 void ff_put_h264_chroma_mc2_neon(uint8_t *, uint8_t *, int, int, int, int); |
8336 | 129 |
130 void ff_avg_h264_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int); | |
131 void ff_avg_h264_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int); | |
10617 | 132 void ff_avg_h264_chroma_mc2_neon(uint8_t *, uint8_t *, int, int, int, int); |
8336 | 133 |
9692 | 134 void ff_vp3_v_loop_filter_neon(uint8_t *, int, int *); |
135 void ff_vp3_h_loop_filter_neon(uint8_t *, int, int *); | |
136 | |
8697 | 137 void ff_vector_fmul_neon(float *dst, const float *src, int len); |
8698 | 138 void ff_vector_fmul_window_neon(float *dst, const float *src0, |
139 const float *src1, const float *win, | |
140 float add_bias, int len); | |
10221 | 141 void ff_vector_fmul_scalar_neon(float *dst, const float *src, float mul, |
142 int len); | |
143 void ff_vector_fmul_sv_scalar_2_neon(float *dst, const float *src, | |
144 const float **vp, float mul, int len); | |
145 void ff_vector_fmul_sv_scalar_4_neon(float *dst, const float *src, | |
146 const float **vp, float mul, int len); | |
147 void ff_sv_fmul_scalar_2_neon(float *dst, const float **vp, float mul, | |
148 int len); | |
149 void ff_sv_fmul_scalar_4_neon(float *dst, const float **vp, float mul, | |
150 int len); | |
151 void ff_butterflies_float_neon(float *v1, float *v2, int len); | |
10228 | 152 float ff_scalarproduct_float_neon(const float *v1, const float *v2, int len); |
10253 | 153 void ff_int32_to_float_fmul_scalar_neon(float *dst, const int *src, |
154 float mul, int len); | |
10274 | 155 void ff_vector_fmul_reverse_neon(float *dst, const float *src0, |
156 const float *src1, int len); | |
10302 | 157 void ff_vector_fmul_add_neon(float *dst, const float *src0, const float *src1, |
158 const float *src2, int len); | |
8697 | 159 |
10276 | 160 void ff_vector_clipf_neon(float *dst, const float *src, float min, float max, |
161 int len); | |
8492 | 162 void ff_float_to_int16_neon(int16_t *, const float *, long); |
163 void ff_float_to_int16_interleave_neon(int16_t *, const float **, long, int); | |
164 | |
10046 | 165 void ff_vorbis_inverse_coupling_neon(float *mag, float *ang, int blocksize); |
166 | |
11243
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
167 int32_t ff_scalarproduct_int16_neon(int16_t *v1, int16_t *v2, int len, |
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
168 int shift); |
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
169 int32_t ff_scalarproduct_and_madd_int16_neon(int16_t *v1, int16_t *v2, |
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
170 int16_t *v3, int len, int mul); |
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
171 |
8334 | 172 void ff_dsputil_init_neon(DSPContext *c, AVCodecContext *avctx) |
173 { | |
10359 | 174 if (!avctx->lowres) { |
175 if (avctx->idct_algo == FF_IDCT_AUTO || | |
176 avctx->idct_algo == FF_IDCT_SIMPLENEON) { | |
10362 | 177 c->idct_put = ff_simple_idct_put_neon; |
178 c->idct_add = ff_simple_idct_add_neon; | |
179 c->idct = ff_simple_idct_neon; | |
10359 | 180 c->idct_permutation_type = FF_PARTTRANS_IDCT_PERM; |
181 } else if ((CONFIG_VP3_DECODER || CONFIG_VP5_DECODER || | |
182 CONFIG_VP6_DECODER) && | |
183 avctx->idct_algo == FF_IDCT_VP3) { | |
10362 | 184 c->idct_put = ff_vp3_idct_put_neon; |
185 c->idct_add = ff_vp3_idct_add_neon; | |
186 c->idct = ff_vp3_idct_neon; | |
10359 | 187 c->idct_permutation_type = FF_TRANSPOSE_IDCT_PERM; |
188 } | |
189 } | |
190 | |
8334 | 191 c->put_pixels_tab[0][0] = ff_put_pixels16_neon; |
192 c->put_pixels_tab[0][1] = ff_put_pixels16_x2_neon; | |
193 c->put_pixels_tab[0][2] = ff_put_pixels16_y2_neon; | |
194 c->put_pixels_tab[0][3] = ff_put_pixels16_xy2_neon; | |
195 c->put_pixels_tab[1][0] = ff_put_pixels8_neon; | |
196 c->put_pixels_tab[1][1] = ff_put_pixels8_x2_neon; | |
197 c->put_pixels_tab[1][2] = ff_put_pixels8_y2_neon; | |
198 c->put_pixels_tab[1][3] = ff_put_pixels8_xy2_neon; | |
199 | |
200 c->put_no_rnd_pixels_tab[0][0] = ff_put_pixels16_neon; | |
201 c->put_no_rnd_pixels_tab[0][1] = ff_put_pixels16_x2_no_rnd_neon; | |
202 c->put_no_rnd_pixels_tab[0][2] = ff_put_pixels16_y2_no_rnd_neon; | |
203 c->put_no_rnd_pixels_tab[0][3] = ff_put_pixels16_xy2_no_rnd_neon; | |
204 c->put_no_rnd_pixels_tab[1][0] = ff_put_pixels8_neon; | |
205 c->put_no_rnd_pixels_tab[1][1] = ff_put_pixels8_x2_no_rnd_neon; | |
206 c->put_no_rnd_pixels_tab[1][2] = ff_put_pixels8_y2_no_rnd_neon; | |
207 c->put_no_rnd_pixels_tab[1][3] = ff_put_pixels8_xy2_no_rnd_neon; | |
208 | |
209 c->avg_pixels_tab[0][0] = ff_avg_pixels16_neon; | |
10375 | 210 c->avg_pixels_tab[1][0] = ff_avg_pixels8_neon; |
8334 | 211 |
9344 | 212 c->add_pixels_clamped = ff_add_pixels_clamped_neon; |
9580 | 213 c->put_pixels_clamped = ff_put_pixels_clamped_neon; |
9345 | 214 c->put_signed_pixels_clamped = ff_put_signed_pixels_clamped_neon; |
9344 | 215 |
9976
e52cd349e708
Only compile in NEON optimizations for H.264 when the H.264 decoder is enabled.
diego
parents:
9975
diff
changeset
|
216 if (CONFIG_H264_DECODER) { |
9977 | 217 c->put_h264_chroma_pixels_tab[0] = ff_put_h264_chroma_mc8_neon; |
218 c->put_h264_chroma_pixels_tab[1] = ff_put_h264_chroma_mc4_neon; | |
10617 | 219 c->put_h264_chroma_pixels_tab[2] = ff_put_h264_chroma_mc2_neon; |
8336 | 220 |
9977 | 221 c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_neon; |
222 c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_neon; | |
10617 | 223 c->avg_h264_chroma_pixels_tab[2] = ff_avg_h264_chroma_mc2_neon; |
8336 | 224 |
9977 | 225 c->put_h264_qpel_pixels_tab[0][ 0] = ff_put_h264_qpel16_mc00_neon; |
226 c->put_h264_qpel_pixels_tab[0][ 1] = ff_put_h264_qpel16_mc10_neon; | |
227 c->put_h264_qpel_pixels_tab[0][ 2] = ff_put_h264_qpel16_mc20_neon; | |
228 c->put_h264_qpel_pixels_tab[0][ 3] = ff_put_h264_qpel16_mc30_neon; | |
229 c->put_h264_qpel_pixels_tab[0][ 4] = ff_put_h264_qpel16_mc01_neon; | |
230 c->put_h264_qpel_pixels_tab[0][ 5] = ff_put_h264_qpel16_mc11_neon; | |
231 c->put_h264_qpel_pixels_tab[0][ 6] = ff_put_h264_qpel16_mc21_neon; | |
232 c->put_h264_qpel_pixels_tab[0][ 7] = ff_put_h264_qpel16_mc31_neon; | |
233 c->put_h264_qpel_pixels_tab[0][ 8] = ff_put_h264_qpel16_mc02_neon; | |
234 c->put_h264_qpel_pixels_tab[0][ 9] = ff_put_h264_qpel16_mc12_neon; | |
235 c->put_h264_qpel_pixels_tab[0][10] = ff_put_h264_qpel16_mc22_neon; | |
236 c->put_h264_qpel_pixels_tab[0][11] = ff_put_h264_qpel16_mc32_neon; | |
237 c->put_h264_qpel_pixels_tab[0][12] = ff_put_h264_qpel16_mc03_neon; | |
238 c->put_h264_qpel_pixels_tab[0][13] = ff_put_h264_qpel16_mc13_neon; | |
239 c->put_h264_qpel_pixels_tab[0][14] = ff_put_h264_qpel16_mc23_neon; | |
240 c->put_h264_qpel_pixels_tab[0][15] = ff_put_h264_qpel16_mc33_neon; | |
8338 | 241 |
9977 | 242 c->put_h264_qpel_pixels_tab[1][ 0] = ff_put_h264_qpel8_mc00_neon; |
243 c->put_h264_qpel_pixels_tab[1][ 1] = ff_put_h264_qpel8_mc10_neon; | |
244 c->put_h264_qpel_pixels_tab[1][ 2] = ff_put_h264_qpel8_mc20_neon; | |
245 c->put_h264_qpel_pixels_tab[1][ 3] = ff_put_h264_qpel8_mc30_neon; | |
246 c->put_h264_qpel_pixels_tab[1][ 4] = ff_put_h264_qpel8_mc01_neon; | |
247 c->put_h264_qpel_pixels_tab[1][ 5] = ff_put_h264_qpel8_mc11_neon; | |
248 c->put_h264_qpel_pixels_tab[1][ 6] = ff_put_h264_qpel8_mc21_neon; | |
249 c->put_h264_qpel_pixels_tab[1][ 7] = ff_put_h264_qpel8_mc31_neon; | |
250 c->put_h264_qpel_pixels_tab[1][ 8] = ff_put_h264_qpel8_mc02_neon; | |
251 c->put_h264_qpel_pixels_tab[1][ 9] = ff_put_h264_qpel8_mc12_neon; | |
252 c->put_h264_qpel_pixels_tab[1][10] = ff_put_h264_qpel8_mc22_neon; | |
253 c->put_h264_qpel_pixels_tab[1][11] = ff_put_h264_qpel8_mc32_neon; | |
254 c->put_h264_qpel_pixels_tab[1][12] = ff_put_h264_qpel8_mc03_neon; | |
255 c->put_h264_qpel_pixels_tab[1][13] = ff_put_h264_qpel8_mc13_neon; | |
256 c->put_h264_qpel_pixels_tab[1][14] = ff_put_h264_qpel8_mc23_neon; | |
257 c->put_h264_qpel_pixels_tab[1][15] = ff_put_h264_qpel8_mc33_neon; | |
8334 | 258 |
9977 | 259 c->avg_h264_qpel_pixels_tab[0][ 0] = ff_avg_h264_qpel16_mc00_neon; |
10616 | 260 c->avg_h264_qpel_pixels_tab[0][ 1] = ff_avg_h264_qpel16_mc10_neon; |
261 c->avg_h264_qpel_pixels_tab[0][ 2] = ff_avg_h264_qpel16_mc20_neon; | |
262 c->avg_h264_qpel_pixels_tab[0][ 3] = ff_avg_h264_qpel16_mc30_neon; | |
263 c->avg_h264_qpel_pixels_tab[0][ 4] = ff_avg_h264_qpel16_mc01_neon; | |
264 c->avg_h264_qpel_pixels_tab[0][ 5] = ff_avg_h264_qpel16_mc11_neon; | |
265 c->avg_h264_qpel_pixels_tab[0][ 6] = ff_avg_h264_qpel16_mc21_neon; | |
266 c->avg_h264_qpel_pixels_tab[0][ 7] = ff_avg_h264_qpel16_mc31_neon; | |
267 c->avg_h264_qpel_pixels_tab[0][ 8] = ff_avg_h264_qpel16_mc02_neon; | |
268 c->avg_h264_qpel_pixels_tab[0][ 9] = ff_avg_h264_qpel16_mc12_neon; | |
269 c->avg_h264_qpel_pixels_tab[0][10] = ff_avg_h264_qpel16_mc22_neon; | |
270 c->avg_h264_qpel_pixels_tab[0][11] = ff_avg_h264_qpel16_mc32_neon; | |
271 c->avg_h264_qpel_pixels_tab[0][12] = ff_avg_h264_qpel16_mc03_neon; | |
272 c->avg_h264_qpel_pixels_tab[0][13] = ff_avg_h264_qpel16_mc13_neon; | |
273 c->avg_h264_qpel_pixels_tab[0][14] = ff_avg_h264_qpel16_mc23_neon; | |
274 c->avg_h264_qpel_pixels_tab[0][15] = ff_avg_h264_qpel16_mc33_neon; | |
8337 | 275 |
10375 | 276 c->avg_h264_qpel_pixels_tab[1][ 0] = ff_avg_h264_qpel8_mc00_neon; |
10616 | 277 c->avg_h264_qpel_pixels_tab[1][ 1] = ff_avg_h264_qpel8_mc10_neon; |
278 c->avg_h264_qpel_pixels_tab[1][ 2] = ff_avg_h264_qpel8_mc20_neon; | |
279 c->avg_h264_qpel_pixels_tab[1][ 3] = ff_avg_h264_qpel8_mc30_neon; | |
280 c->avg_h264_qpel_pixels_tab[1][ 4] = ff_avg_h264_qpel8_mc01_neon; | |
281 c->avg_h264_qpel_pixels_tab[1][ 5] = ff_avg_h264_qpel8_mc11_neon; | |
282 c->avg_h264_qpel_pixels_tab[1][ 6] = ff_avg_h264_qpel8_mc21_neon; | |
283 c->avg_h264_qpel_pixels_tab[1][ 7] = ff_avg_h264_qpel8_mc31_neon; | |
284 c->avg_h264_qpel_pixels_tab[1][ 8] = ff_avg_h264_qpel8_mc02_neon; | |
285 c->avg_h264_qpel_pixels_tab[1][ 9] = ff_avg_h264_qpel8_mc12_neon; | |
286 c->avg_h264_qpel_pixels_tab[1][10] = ff_avg_h264_qpel8_mc22_neon; | |
287 c->avg_h264_qpel_pixels_tab[1][11] = ff_avg_h264_qpel8_mc32_neon; | |
288 c->avg_h264_qpel_pixels_tab[1][12] = ff_avg_h264_qpel8_mc03_neon; | |
289 c->avg_h264_qpel_pixels_tab[1][13] = ff_avg_h264_qpel8_mc13_neon; | |
290 c->avg_h264_qpel_pixels_tab[1][14] = ff_avg_h264_qpel8_mc23_neon; | |
291 c->avg_h264_qpel_pixels_tab[1][15] = ff_avg_h264_qpel8_mc33_neon; | |
9976
e52cd349e708
Only compile in NEON optimizations for H.264 when the H.264 decoder is enabled.
diego
parents:
9975
diff
changeset
|
292 } |
8492 | 293 |
9975
d6d7e8d4a04d
Do not redundantly check for both CONFIG_THEORA_DECODER and CONFIG_VP3_DECODER.
diego
parents:
9692
diff
changeset
|
294 if (CONFIG_VP3_DECODER) { |
9692 | 295 c->vp3_v_loop_filter = ff_vp3_v_loop_filter_neon; |
296 c->vp3_h_loop_filter = ff_vp3_h_loop_filter_neon; | |
297 } | |
298 | |
10362 | 299 c->vector_fmul = ff_vector_fmul_neon; |
300 c->vector_fmul_window = ff_vector_fmul_window_neon; | |
301 c->vector_fmul_scalar = ff_vector_fmul_scalar_neon; | |
302 c->butterflies_float = ff_butterflies_float_neon; | |
303 c->scalarproduct_float = ff_scalarproduct_float_neon; | |
10253 | 304 c->int32_to_float_fmul_scalar = ff_int32_to_float_fmul_scalar_neon; |
10362 | 305 c->vector_fmul_reverse = ff_vector_fmul_reverse_neon; |
306 c->vector_fmul_add = ff_vector_fmul_add_neon; | |
307 c->vector_clipf = ff_vector_clipf_neon; | |
10221 | 308 |
309 c->vector_fmul_sv_scalar[0] = ff_vector_fmul_sv_scalar_2_neon; | |
310 c->vector_fmul_sv_scalar[1] = ff_vector_fmul_sv_scalar_4_neon; | |
311 | |
312 c->sv_fmul_scalar[0] = ff_sv_fmul_scalar_2_neon; | |
313 c->sv_fmul_scalar[1] = ff_sv_fmul_scalar_4_neon; | |
8697 | 314 |
8492 | 315 if (!(avctx->flags & CODEC_FLAG_BITEXACT)) { |
10362 | 316 c->float_to_int16 = ff_float_to_int16_neon; |
8492 | 317 c->float_to_int16_interleave = ff_float_to_int16_interleave_neon; |
318 } | |
10046 | 319 |
320 if (CONFIG_VORBIS_DECODER) | |
321 c->vorbis_inverse_coupling = ff_vorbis_inverse_coupling_neon; | |
11243
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
322 |
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
323 c->scalarproduct_int16 = ff_scalarproduct_int16_neon; |
e71b0be9ac79
ARM: NEON scalarproduct_int16 and scalarproduct_and_madd_int16
mru
parents:
10617
diff
changeset
|
324 c->scalarproduct_and_madd_int16 = ff_scalarproduct_and_madd_int16_neon; |
8334 | 325 } |