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 "config.h"
20
21 #include "libavutil/attributes.h"
22 #include "libavutil/cpu.h"
23 #include "libavutil/float_dsp.h"
24 #include "cpu.h"
25 #include "asm.h"
26
27 void ff_vector_fmul_sse(float *dst, const float *src0, const float *src1,
28 int len);
29 void ff_vector_fmul_avx(float *dst, const float *src0, const float *src1,
30 int len);
31
32 void ff_vector_fmac_scalar_sse(float *dst, const float *src, float mul,
33 int len);
34 void ff_vector_fmac_scalar_avx(float *dst, const float *src, float mul,
35 int len);
36 void ff_vector_fmac_scalar_fma3(float *dst, const float *src, float mul,
37 int len);
38
39 void ff_vector_fmul_scalar_sse(float *dst, const float *src, float mul,
40 int len);
41
42 void ff_vector_dmul_scalar_sse2(double *dst, const double *src,
43 double mul, int len);
44 void ff_vector_dmul_scalar_avx(double *dst, const double *src,
45 double mul, int len);
46
47 void ff_vector_fmul_window_3dnowext(float *dst, const float *src0,
48 const float *src1, const float *win, int len);
49 void ff_vector_fmul_window_sse(float *dst, const float *src0,
50 const float *src1, const float *win, int len);
51
52 void ff_vector_fmul_add_sse(float *dst, const float *src0, const float *src1,
53 const float *src2, int len);
54 void ff_vector_fmul_add_avx(float *dst, const float *src0, const float *src1,
55 const float *src2, int len);
56 void ff_vector_fmul_add_fma3(float *dst, const float *src0, const float *src1,
57 const float *src2, int len);
58
59 void ff_vector_fmul_reverse_sse(float *dst, const float *src0,
60 const float *src1, int len);
61 void ff_vector_fmul_reverse_avx(float *dst, const float *src0,
62 const float *src1, int len);
63
64 float ff_scalarproduct_float_sse(const float *v1, const float *v2, int order);
65
66 void ff_butterflies_float_sse(float *src0, float *src1, int len);
67
ff_float_dsp_init_x86(AVFloatDSPContext * fdsp)68 av_cold void ff_float_dsp_init_x86(AVFloatDSPContext *fdsp)
69 {
70 int cpu_flags = av_get_cpu_flags();
71
72 #if (HAVE_AMD3DNOWEXT_EXTERNAL == 1)
73 if (EXTERNAL_AMD3DNOWEXT(cpu_flags)) {
74 fdsp->vector_fmul_window = ff_vector_fmul_window_3dnowext;
75 }
76 #endif
77 #if (HAVE_SSE_EXTERNAL == 1)
78 if (EXTERNAL_SSE(cpu_flags)) {
79 fdsp->vector_fmul = ff_vector_fmul_sse;
80 fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_sse;
81 fdsp->vector_fmul_scalar = ff_vector_fmul_scalar_sse;
82 fdsp->vector_fmul_window = ff_vector_fmul_window_sse;
83 fdsp->vector_fmul_add = ff_vector_fmul_add_sse;
84 fdsp->vector_fmul_reverse = ff_vector_fmul_reverse_sse;
85 fdsp->scalarproduct_float = ff_scalarproduct_float_sse;
86 fdsp->butterflies_float = ff_butterflies_float_sse;
87 }
88 #endif
89 #if (HAVE_SSE2_EXTERNAL == 1)
90 if (EXTERNAL_SSE2(cpu_flags)) {
91 fdsp->vector_dmul_scalar = ff_vector_dmul_scalar_sse2;
92 }
93 #endif
94 #if (HAVE_AVX_EXTERNAL == 1)
95 if (EXTERNAL_AVX(cpu_flags)) {
96 fdsp->vector_fmul = ff_vector_fmul_avx;
97 fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_avx;
98 fdsp->vector_dmul_scalar = ff_vector_dmul_scalar_avx;
99 fdsp->vector_fmul_add = ff_vector_fmul_add_avx;
100 fdsp->vector_fmul_reverse = ff_vector_fmul_reverse_avx;
101 }
102 #endif
103 #if (HAVE_FMA3_EXTERNAL == 1)
104 if (EXTERNAL_FMA3(cpu_flags)) {
105 fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_fma3;
106 fdsp->vector_fmul_add = ff_vector_fmul_add_fma3;
107 }
108 #endif
109 }
110