1;******************************************************************************
2;* x86 optimized Format Conversion Utils
3;* Copyright (c) 2008 Loren Merritt
4;*
5;* This file is part of FFmpeg.
6;*
7;* FFmpeg is free software; you can redistribute it and/or
8;* modify it under the terms of the GNU Lesser General Public
9;* License as published by the Free Software Foundation; either
10;* version 2.1 of the License, or (at your option) any later version.
11;*
12;* FFmpeg is distributed in the hope that it will be useful,
13;* but WITHOUT ANY WARRANTY; without even the implied warranty of
14;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
15;* Lesser General Public License for more details.
16;*
17;* You should have received a copy of the GNU Lesser General Public
18;* License along with FFmpeg; if not, write to the Free Software
19;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20;******************************************************************************
21
22%include "libavutil/x86/x86util.asm"
23
24SECTION .text
25
26;------------------------------------------------------------------------------
27; void ff_int32_to_float_fmul_scalar(float *dst, const int32_t *src, float mul,
28;                                    int len);
29;------------------------------------------------------------------------------
30%macro INT32_TO_FLOAT_FMUL_SCALAR 1
31%if UNIX64
32cglobal int32_to_float_fmul_scalar, 3, 3, %1, dst, src, len
33%else
34cglobal int32_to_float_fmul_scalar, 4, 4, %1, dst, src, mul, len
35%endif
36%if WIN64
37    SWAP 0, 2
38%elif ARCH_X86_32
39    movss   m0, mulm
40%endif
41    SPLATD  m0
42    shl     lend, 2
43    add     srcq, lenq
44    add     dstq, lenq
45    neg     lenq
46.loop:
47%if cpuflag(sse2)
48    cvtdq2ps  m1, [srcq+lenq   ]
49    cvtdq2ps  m2, [srcq+lenq+16]
50%else
51    cvtpi2ps  m1, [srcq+lenq   ]
52    cvtpi2ps  m3, [srcq+lenq+ 8]
53    cvtpi2ps  m2, [srcq+lenq+16]
54    cvtpi2ps  m4, [srcq+lenq+24]
55    movlhps   m1, m3
56    movlhps   m2, m4
57%endif
58    mulps     m1, m0
59    mulps     m2, m0
60    mova  [dstq+lenq   ], m1
61    mova  [dstq+lenq+16], m2
62    add     lenq, 32
63    jl .loop
64%if notcpuflag(sse2)
65    ;; cvtpi2ps switches to MMX even if the source is a memory location
66    ;; possible an error in documentation since every tested CPU disagrees with
67    ;; that. Use emms anyway since the vast majority of machines will use the
68    ;; SSE2 variant
69    emms
70%endif
71    RET
72%endmacro
73
74INIT_XMM sse
75INT32_TO_FLOAT_FMUL_SCALAR 5
76INIT_XMM sse2
77INT32_TO_FLOAT_FMUL_SCALAR 3
78
79;------------------------------------------------------------------------------
80; void ff_int32_to_float_fmul_array8(FmtConvertContext *c, float *dst, const int32_t *src,
81;                                    const float *mul, int len);
82;------------------------------------------------------------------------------
83%macro INT32_TO_FLOAT_FMUL_ARRAY8 0
84cglobal int32_to_float_fmul_array8, 5, 5, 5, c, dst, src, mul, len
85    shl     lend, 2
86    add     srcq, lenq
87    add     dstq, lenq
88    neg     lenq
89.loop:
90    movss     m0, [mulq]
91    SPLATD    m0
92%if cpuflag(sse2)
93    cvtdq2ps  m1, [srcq+lenq   ]
94    cvtdq2ps  m2, [srcq+lenq+16]
95%else
96    cvtpi2ps  m1, [srcq+lenq   ]
97    cvtpi2ps  m3, [srcq+lenq+ 8]
98    cvtpi2ps  m2, [srcq+lenq+16]
99    cvtpi2ps  m4, [srcq+lenq+24]
100    movlhps   m1, m3
101    movlhps   m2, m4
102%endif
103    mulps     m1, m0
104    mulps     m2, m0
105    mova  [dstq+lenq   ], m1
106    mova  [dstq+lenq+16], m2
107    add     mulq, 4
108    add     lenq, 32
109    jl .loop
110%if notcpuflag(sse2)
111    ;; cvtpi2ps switches to MMX even if the source is a memory location
112    ;; possible an error in documentation since every tested CPU disagrees with
113    ;; that. Use emms anyway since the vast majority of machines will use the
114    ;; SSE2 variant
115    emms
116%endif
117    RET
118%endmacro
119
120INIT_XMM sse
121INT32_TO_FLOAT_FMUL_ARRAY8
122INIT_XMM sse2
123INT32_TO_FLOAT_FMUL_ARRAY8
124
125