Go to most recent revision | Details | Last modification | View Log | RSS feed
Rev | Author | Line No. | Line |
---|---|---|---|
4349 | Serge | 1 | /* |
2 | * This file is part of FFmpeg. |
||
3 | * |
||
4 | * FFmpeg is free software; you can redistribute it and/or |
||
5 | * modify it under the terms of the GNU Lesser General Public |
||
6 | * License as published by the Free Software Foundation; either |
||
7 | * version 2.1 of the License, or (at your option) any later version. |
||
8 | * |
||
9 | * FFmpeg is distributed in the hope that it will be useful, |
||
10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
||
11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
||
12 | * Lesser General Public License for more details. |
||
13 | * |
||
14 | * You should have received a copy of the GNU Lesser General Public |
||
15 | * License along with FFmpeg; if not, write to the Free Software |
||
16 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
||
17 | */ |
||
18 | |||
19 | #include |
||
20 | |||
21 | #include "config.h" |
||
22 | #include "libavutil/attributes.h" |
||
23 | #include "libavutil/cpu.h" |
||
24 | #include "libavutil/x86/cpu.h" |
||
25 | #include "libavcodec/h264chroma.h" |
||
26 | |||
27 | void ff_put_h264_chroma_mc8_rnd_mmx (uint8_t *dst, uint8_t *src, |
||
28 | int stride, int h, int x, int y); |
||
29 | void ff_avg_h264_chroma_mc8_rnd_mmxext(uint8_t *dst, uint8_t *src, |
||
30 | int stride, int h, int x, int y); |
||
31 | void ff_avg_h264_chroma_mc8_rnd_3dnow(uint8_t *dst, uint8_t *src, |
||
32 | int stride, int h, int x, int y); |
||
33 | |||
34 | void ff_put_h264_chroma_mc4_mmx (uint8_t *dst, uint8_t *src, |
||
35 | int stride, int h, int x, int y); |
||
36 | void ff_avg_h264_chroma_mc4_mmxext (uint8_t *dst, uint8_t *src, |
||
37 | int stride, int h, int x, int y); |
||
38 | void ff_avg_h264_chroma_mc4_3dnow (uint8_t *dst, uint8_t *src, |
||
39 | int stride, int h, int x, int y); |
||
40 | |||
41 | void ff_put_h264_chroma_mc2_mmxext (uint8_t *dst, uint8_t *src, |
||
42 | int stride, int h, int x, int y); |
||
43 | void ff_avg_h264_chroma_mc2_mmxext (uint8_t *dst, uint8_t *src, |
||
44 | int stride, int h, int x, int y); |
||
45 | |||
46 | void ff_put_h264_chroma_mc8_rnd_ssse3(uint8_t *dst, uint8_t *src, |
||
47 | int stride, int h, int x, int y); |
||
48 | void ff_put_h264_chroma_mc4_ssse3 (uint8_t *dst, uint8_t *src, |
||
49 | int stride, int h, int x, int y); |
||
50 | |||
51 | void ff_avg_h264_chroma_mc8_rnd_ssse3(uint8_t *dst, uint8_t *src, |
||
52 | int stride, int h, int x, int y); |
||
53 | void ff_avg_h264_chroma_mc4_ssse3 (uint8_t *dst, uint8_t *src, |
||
54 | int stride, int h, int x, int y); |
||
55 | |||
56 | #define CHROMA_MC(OP, NUM, DEPTH, OPT) \ |
||
57 | void ff_ ## OP ## _h264_chroma_mc ## NUM ## _ ## DEPTH ## _ ## OPT \ |
||
58 | (uint8_t *dst, uint8_t *src, \ |
||
59 | int stride, int h, int x, int y); |
||
60 | |||
61 | CHROMA_MC(put, 2, 10, mmxext) |
||
62 | CHROMA_MC(avg, 2, 10, mmxext) |
||
63 | CHROMA_MC(put, 4, 10, mmxext) |
||
64 | CHROMA_MC(avg, 4, 10, mmxext) |
||
65 | CHROMA_MC(put, 8, 10, sse2) |
||
66 | CHROMA_MC(avg, 8, 10, sse2) |
||
67 | CHROMA_MC(put, 8, 10, avx) |
||
68 | CHROMA_MC(avg, 8, 10, avx) |
||
69 | |||
70 | av_cold void ff_h264chroma_init_x86(H264ChromaContext *c, int bit_depth) |
||
71 | { |
||
72 | #if HAVE_YASM |
||
73 | int high_bit_depth = bit_depth > 8; |
||
74 | int cpu_flags = av_get_cpu_flags(); |
||
75 | |||
76 | if (EXTERNAL_MMX(cpu_flags) && !high_bit_depth) { |
||
77 | c->put_h264_chroma_pixels_tab[0] = ff_put_h264_chroma_mc8_rnd_mmx; |
||
78 | c->put_h264_chroma_pixels_tab[1] = ff_put_h264_chroma_mc4_mmx; |
||
79 | } |
||
80 | |||
81 | if (EXTERNAL_AMD3DNOW(cpu_flags) && !high_bit_depth) { |
||
82 | c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_rnd_3dnow; |
||
83 | c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_3dnow; |
||
84 | } |
||
85 | |||
86 | if (EXTERNAL_MMXEXT(cpu_flags) && !high_bit_depth) { |
||
87 | c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_rnd_mmxext; |
||
88 | c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_mmxext; |
||
89 | c->avg_h264_chroma_pixels_tab[2] = ff_avg_h264_chroma_mc2_mmxext; |
||
90 | c->put_h264_chroma_pixels_tab[2] = ff_put_h264_chroma_mc2_mmxext; |
||
91 | } |
||
92 | |||
93 | if (EXTERNAL_MMXEXT(cpu_flags) && bit_depth > 8 && bit_depth <= 10) { |
||
94 | c->put_h264_chroma_pixels_tab[2] = ff_put_h264_chroma_mc2_10_mmxext; |
||
95 | c->avg_h264_chroma_pixels_tab[2] = ff_avg_h264_chroma_mc2_10_mmxext; |
||
96 | c->put_h264_chroma_pixels_tab[1] = ff_put_h264_chroma_mc4_10_mmxext; |
||
97 | c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_10_mmxext; |
||
98 | } |
||
99 | |||
100 | if (EXTERNAL_SSE2(cpu_flags) && bit_depth > 8 && bit_depth <= 10) { |
||
101 | c->put_h264_chroma_pixels_tab[0] = ff_put_h264_chroma_mc8_10_sse2; |
||
102 | c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_10_sse2; |
||
103 | } |
||
104 | |||
105 | if (EXTERNAL_SSSE3(cpu_flags) && !high_bit_depth) { |
||
106 | c->put_h264_chroma_pixels_tab[0] = ff_put_h264_chroma_mc8_rnd_ssse3; |
||
107 | c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_rnd_ssse3; |
||
108 | c->put_h264_chroma_pixels_tab[1] = ff_put_h264_chroma_mc4_ssse3; |
||
109 | c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_ssse3; |
||
110 | } |
||
111 | |||
112 | if (EXTERNAL_AVX(cpu_flags) && bit_depth > 8 && bit_depth <= 10) { |
||
113 | // AVX implies !cache64. |
||
114 | // TODO: Port cache(32|64) detection from x264. |
||
115 | c->put_h264_chroma_pixels_tab[0] = ff_put_h264_chroma_mc8_10_avx; |
||
116 | c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_10_avx; |
||
117 | } |
||
118 | #endif |
||
119 | }=>=>=> |