2 * VC-1 and WMV3 - DSP functions MMX-optimized
3 * Copyright (c) 2007 Christophe GISQUET <christophe.gisquet@free.fr>
5 * Permission is hereby granted, free of charge, to any person
6 * obtaining a copy of this software and associated documentation
7 * files (the "Software"), to deal in the Software without
8 * restriction, including without limitation the rights to use,
9 * copy, modify, merge, publish, distribute, sublicense, and/or sell
10 * copies of the Software, and to permit persons to whom the
11 * Software is furnished to do so, subject to the following
14 * The above copyright notice and this permission notice shall be
15 * included in all copies or substantial portions of the Software.
17 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
18 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
19 * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
20 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
21 * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
22 * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
23 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
24 * OTHER DEALINGS IN THE SOFTWARE.
27 #include "libavutil/attributes.h"
28 #include "libavutil/cpu.h"
29 #include "libavutil/x86/cpu.h"
30 #include "libavcodec/vc1dsp.h"
31 #include "dsputil_x86.h"
35 #define LOOP_FILTER(EXT) \
36 void ff_vc1_v_loop_filter4_ ## EXT(uint8_t *src, int stride, int pq); \
37 void ff_vc1_h_loop_filter4_ ## EXT(uint8_t *src, int stride, int pq); \
38 void ff_vc1_v_loop_filter8_ ## EXT(uint8_t *src, int stride, int pq); \
39 void ff_vc1_h_loop_filter8_ ## EXT(uint8_t *src, int stride, int pq); \
41 static void vc1_v_loop_filter16_ ## EXT(uint8_t *src, int stride, int pq) \
43 ff_vc1_v_loop_filter8_ ## EXT(src, stride, pq); \
44 ff_vc1_v_loop_filter8_ ## EXT(src+8, stride, pq); \
47 static void vc1_h_loop_filter16_ ## EXT(uint8_t *src, int stride, int pq) \
49 ff_vc1_h_loop_filter8_ ## EXT(src, stride, pq); \
50 ff_vc1_h_loop_filter8_ ## EXT(src+8*stride, stride, pq); \
58 void ff_vc1_h_loop_filter8_sse4(uint8_t *src, int stride, int pq);
60 static void vc1_h_loop_filter16_sse4(uint8_t *src, int stride, int pq)
62 ff_vc1_h_loop_filter8_sse4(src, stride, pq);
63 ff_vc1_h_loop_filter8_sse4(src+8*stride, stride, pq);
66 static void avg_vc1_mspel_mc00_mmxext(uint8_t *dst, const uint8_t *src,
67 ptrdiff_t stride, int rnd)
69 ff_avg_pixels8_mmxext(dst, src, stride, 8);
71 #endif /* HAVE_YASM */
73 void ff_put_vc1_chroma_mc8_nornd_mmx (uint8_t *dst, uint8_t *src,
74 int stride, int h, int x, int y);
75 void ff_avg_vc1_chroma_mc8_nornd_mmxext(uint8_t *dst, uint8_t *src,
76 int stride, int h, int x, int y);
77 void ff_avg_vc1_chroma_mc8_nornd_3dnow(uint8_t *dst, uint8_t *src,
78 int stride, int h, int x, int y);
79 void ff_put_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, uint8_t *src,
80 int stride, int h, int x, int y);
81 void ff_avg_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, uint8_t *src,
82 int stride, int h, int x, int y);
85 av_cold void ff_vc1dsp_init_x86(VC1DSPContext *dsp)
87 int cpu_flags = av_get_cpu_flags();
89 if (INLINE_MMX(cpu_flags))
90 ff_vc1dsp_init_mmx(dsp);
92 if (INLINE_MMXEXT(cpu_flags))
93 ff_vc1dsp_init_mmxext(dsp);
95 #define ASSIGN_LF(EXT) \
96 dsp->vc1_v_loop_filter4 = ff_vc1_v_loop_filter4_ ## EXT; \
97 dsp->vc1_h_loop_filter4 = ff_vc1_h_loop_filter4_ ## EXT; \
98 dsp->vc1_v_loop_filter8 = ff_vc1_v_loop_filter8_ ## EXT; \
99 dsp->vc1_h_loop_filter8 = ff_vc1_h_loop_filter8_ ## EXT; \
100 dsp->vc1_v_loop_filter16 = vc1_v_loop_filter16_ ## EXT; \
101 dsp->vc1_h_loop_filter16 = vc1_h_loop_filter16_ ## EXT
104 if (EXTERNAL_MMX(cpu_flags)) {
105 dsp->put_no_rnd_vc1_chroma_pixels_tab[0] = ff_put_vc1_chroma_mc8_nornd_mmx;
107 if (EXTERNAL_AMD3DNOW(cpu_flags)) {
108 dsp->avg_no_rnd_vc1_chroma_pixels_tab[0] = ff_avg_vc1_chroma_mc8_nornd_3dnow;
110 if (EXTERNAL_MMXEXT(cpu_flags)) {
112 dsp->avg_no_rnd_vc1_chroma_pixels_tab[0] = ff_avg_vc1_chroma_mc8_nornd_mmxext;
114 dsp->avg_vc1_mspel_pixels_tab[0] = avg_vc1_mspel_mc00_mmxext;
116 if (EXTERNAL_SSE2(cpu_flags)) {
117 dsp->vc1_v_loop_filter8 = ff_vc1_v_loop_filter8_sse2;
118 dsp->vc1_h_loop_filter8 = ff_vc1_h_loop_filter8_sse2;
119 dsp->vc1_v_loop_filter16 = vc1_v_loop_filter16_sse2;
120 dsp->vc1_h_loop_filter16 = vc1_h_loop_filter16_sse2;
122 if (EXTERNAL_SSSE3(cpu_flags)) {
124 dsp->put_no_rnd_vc1_chroma_pixels_tab[0] = ff_put_vc1_chroma_mc8_nornd_ssse3;
125 dsp->avg_no_rnd_vc1_chroma_pixels_tab[0] = ff_avg_vc1_chroma_mc8_nornd_ssse3;
127 if (EXTERNAL_SSE4(cpu_flags)) {
128 dsp->vc1_h_loop_filter8 = ff_vc1_h_loop_filter8_sse4;
129 dsp->vc1_h_loop_filter16 = vc1_h_loop_filter16_sse4;
131 #endif /* HAVE_YASM */