1 ;*****************************************************************************
2 ;* x86-optimized Float DSP functions
4 ;* Copyright 2006 Loren Merritt
6 ;* This file is part of FFmpeg.
8 ;* FFmpeg is free software; you can redistribute it and/or
9 ;* modify it under the terms of the GNU Lesser General Public
10 ;* License as published by the Free Software Foundation; either
11 ;* version 2.1 of the License, or (at your option) any later version.
13 ;* FFmpeg is distributed in the hope that it will be useful,
14 ;* but WITHOUT ANY WARRANTY; without even the implied warranty of
15 ;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 ;* Lesser General Public License for more details.
18 ;* You should have received a copy of the GNU Lesser General Public
19 ;* License along with FFmpeg; if not, write to the Free Software
20 ;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 ;******************************************************************************
23 %include "x86util.asm"
27 ;-----------------------------------------------------------------------------
28 ; void vector_fmul(float *dst, const float *src0, const float *src1, int len)
29 ;-----------------------------------------------------------------------------
31 cglobal vector_fmul, 4,4,2, dst, src0, src1, len
32 lea lenq, [lend*4 - 2*mmsize]
35 mova m0, [src0q + lenq]
36 mova m1, [src0q + lenq + mmsize]
37 mulps m0, m0, [src1q + lenq]
38 mulps m1, m1, [src1q + lenq + mmsize]
39 mova [dstq + lenq], m0
40 mova [dstq + lenq + mmsize], m1
54 ;------------------------------------------------------------------------------
55 ; void ff_vector_fmac_scalar(float *dst, const float *src, float mul, int len)
56 ;------------------------------------------------------------------------------
58 %macro VECTOR_FMAC_SCALAR 0
60 cglobal vector_fmac_scalar, 3,3,3, dst, src, len
62 cglobal vector_fmac_scalar, 4,4,3, dst, src, mul, len
72 vinsertf128 m0, m0, xmm0, 1
75 lea lenq, [lend*4-2*mmsize]
77 mulps m1, m0, [srcq+lenq ]
78 mulps m2, m0, [srcq+lenq+mmsize]
79 addps m1, m1, [dstq+lenq ]
80 addps m2, m2, [dstq+lenq+mmsize]
82 mova [dstq+lenq+mmsize], m2
95 ;------------------------------------------------------------------------------
96 ; void ff_vector_fmul_scalar(float *dst, const float *src, float mul, int len)
97 ;------------------------------------------------------------------------------
99 %macro VECTOR_FMUL_SCALAR 0
101 cglobal vector_fmul_scalar, 3,3,2, dst, src, len
103 cglobal vector_fmul_scalar, 4,4,3, dst, src, mul, len
111 lea lenq, [lend*4-mmsize]
124 ;------------------------------------------------------------------------------
125 ; void ff_vector_dmul_scalar(double *dst, const double *src, double mul,
127 ;------------------------------------------------------------------------------
129 %macro VECTOR_DMUL_SCALAR 0
131 cglobal vector_dmul_scalar, 3,3,3, dst, src, len
133 cglobal vector_dmul_scalar, 4,4,3, dst, src, mul, len
136 VBROADCASTSD m0, mulm
141 vinsertf128 ymm2, ymm2, xmm2, 1
147 vinsertf128 ymm0, ymm0, xmm0, 1
151 lea lenq, [lend*8-2*mmsize]
153 mulpd m1, m0, [srcq+lenq ]
154 mulpd m2, m0, [srcq+lenq+mmsize]
155 mova [dstq+lenq ], m1
156 mova [dstq+lenq+mmsize], m2
164 %if HAVE_AVX_EXTERNAL