float_dsp.asm 2.74 KB
Newer Older
1 2 3
;*****************************************************************************
;* x86-optimized Float DSP functions
;*
4
;* Copyright 2006 Loren Merritt
5
;*
6 7 8
;* This file is part of FFmpeg.
;*
;* FFmpeg is free software; you can redistribute it and/or
9 10 11 12
;* modify it under the terms of the GNU Lesser General Public
;* License as published by the Free Software Foundation; either
;* 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 15 16 17 18
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
;* Lesser General Public License for more details.
;*
;* 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 21 22 23
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************

%include "x86inc.asm"
24
%include "x86util.asm"
25 26 27 28 29 30 31 32 33 34

SECTION .text

;-----------------------------------------------------------------------------
; void vector_fmul(float *dst, const float *src0, const float *src1, int len)
;-----------------------------------------------------------------------------
%macro VECTOR_FMUL 0
cglobal vector_fmul, 4,4,2, dst, src0, src1, len
    lea       lenq, [lend*4 - 2*mmsize]
ALIGN 16
35
.loop:
36 37 38 39 40 41 42 43 44 45 46 47 48 49
    mova      m0,   [src0q + lenq]
    mova      m1,   [src0q + lenq + mmsize]
    mulps     m0, m0, [src1q + lenq]
    mulps     m1, m1, [src1q + lenq + mmsize]
    mova      [dstq + lenq], m0
    mova      [dstq + lenq + mmsize], m1

    sub       lenq, 2*mmsize
    jge       .loop
    REP_RET
%endmacro

INIT_XMM sse
VECTOR_FMUL
50
%if HAVE_AVX_EXTERNAL
51 52 53
INIT_YMM avx
VECTOR_FMUL
%endif
54 55 56 57 58 59 60 61 62 63 64 65 66 67

;------------------------------------------------------------------------------
; void ff_vector_fmac_scalar(float *dst, const float *src, float mul, int len)
;------------------------------------------------------------------------------

%macro VECTOR_FMAC_SCALAR 0
%if UNIX64
cglobal vector_fmac_scalar, 3,3,3, dst, src, len
%else
cglobal vector_fmac_scalar, 4,4,3, dst, src, mul, len
%endif
%if ARCH_X86_32
    VBROADCASTSS m0, mulm
%else
68 69 70
%if WIN64
    mova       xmm0, xmm2
%endif
71 72 73 74 75 76
    shufps     xmm0, xmm0, 0
%if cpuflag(avx)
    vinsertf128  m0, m0, xmm0, 1
%endif
%endif
    lea    lenq, [lend*4-2*mmsize]
77
.loop:
78 79 80 81 82 83 84 85 86 87 88 89 90
    mulps    m1, m0, [srcq+lenq       ]
    mulps    m2, m0, [srcq+lenq+mmsize]
    addps    m1, m1, [dstq+lenq       ]
    addps    m2, m2, [dstq+lenq+mmsize]
    mova  [dstq+lenq       ], m1
    mova  [dstq+lenq+mmsize], m2
    sub    lenq, 2*mmsize
    jge .loop
    REP_RET
%endmacro

INIT_XMM sse
VECTOR_FMAC_SCALAR
91
%if HAVE_AVX_EXTERNAL
92 93 94
INIT_YMM avx
VECTOR_FMAC_SCALAR
%endif