1
/* Copyright (C) 2003 Jean-Marc Valin */
4
@brief ARM-tuned fixed-point operations
7
Redistribution and use in source and binary forms, with or without
8
modification, are permitted provided that the following conditions
11
- Redistributions of source code must retain the above copyright
12
notice, this list of conditions and the following disclaimer.
14
- Redistributions in binary form must reproduce the above copyright
15
notice, this list of conditions and the following disclaimer in the
16
documentation and/or other materials provided with the distribution.
18
- Neither the name of the Xiph.org Foundation nor the names of its
19
contributors may be used to endorse or promote products derived from
20
this software without specific prior written permission.
22
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
23
``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
24
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
25
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR
26
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
27
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
28
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
29
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
30
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
31
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
32
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
38
#define SHR(a,shift) ((a) >> (shift))
39
#define SHL(a,shift) ((a) << (shift))
41
#define SATURATE(x,a) ((x)>(a) ? (a) : (x)<-(a) ? -(a) : (x))
44
#define ADD16(a,b) ((short)((short)(a)+(short)(b)))
45
#define SUB16(a,b) ((a)-(b))
46
#define ADD32(a,b) ((a)+(b))
47
#define SUB32(a,b) ((a)-(b))
48
#define ADD64(a,b) ((a)+(b))
50
#define PSHR(a,shift) (SHR((a)+(1<<((shift)-1)),shift))
52
/* result fits in 16 bits */
53
#define MULT16_16_16(a,b) (((short)(a))*((short)(b)))
55
static inline spx_word32_t MULT16_16(spx_word16_t x, spx_word16_t y) {
57
asm ("smulbb %0,%1,%2;\n"
63
static inline spx_word32_t MAC16_16(spx_word32_t a, spx_word16_t x, spx_word32_t y) {
65
asm ("smlabb %0,%1,%2,%3;\n"
67
: "%r"(x),"r"(y),"r"(a));
71
#define MULT16_32_Q12(a,b) ADD32(MULT16_16((a),SHR((b),12)), SHR(MULT16_16((a),((b)&0x00000fff)),12))
72
#define MULT16_32_Q13(a,b) ADD32(MULT16_16((a),SHR((b),13)), SHR(MULT16_16((a),((b)&0x00001fff)),13))
73
#define MULT16_32_Q14(a,b) ADD32(MULT16_16((a),SHR((b),14)), SHR(MULT16_16((a),((b)&0x00003fff)),14))
75
static inline spx_word32_t MULT16_32_Q15(spx_word16_t x, spx_word32_t y) {
77
asm ("smulwb %0,%1,%2;\n"
82
static inline spx_word32_t MAC16_32_Q15(spx_word32_t a, spx_word16_t x, spx_word32_t y) {
84
asm ("smlawb %0,%1,%2,%3;\n"
86
: "%r"(y<<1),"r"(x),"r"(a));
89
static inline spx_word32_t MULT16_32_Q11(spx_word16_t x, spx_word32_t y) {
91
asm ("smulwb %0,%1,%2;\n"
96
static inline spx_word32_t MAC16_32_Q11(spx_word32_t a, spx_word16_t x, spx_word32_t y) {
98
asm ("smlawb %0,%1,%2,%3;\n"
100
: "%r"(y<<5),"r"(x),"r"(a));
104
#define MAC16_16_Q11(c,a,b) (ADD32((c),SHR(MULT16_16((a),(b)),11)))
106
#define MULT16_16_Q11(a,b) (SHR(MULT16_16((a),(b)),11))
107
#define MULT16_16_Q13(a,b) (SHR(MULT16_16((a),(b)),13))
108
#define MULT16_16_Q14(a,b) (SHR(MULT16_16((a),(b)),14))
109
#define MULT16_16_Q15(a,b) (SHR(MULT16_16((a),(b)),15))
111
#define MULT16_16_P13(a,b) (SHR(ADD32(4096,MULT16_16((a),(b))),13))
112
#define MULT16_16_P14(a,b) (SHR(ADD32(8192,MULT16_16((a),(b))),14))
113
#define MULT16_16_P15(a,b) (SHR(ADD32(16384,MULT16_16((a),(b))),15))
115
#define MUL_16_32_R15(a,bh,bl) ADD32(MULT16_16((a),(bh)), SHR(MULT16_16((a),(bl)),15))
119
#define DIV32_16(a,b) ((short)(((signed int)(a))/((short)(b))))
120
#define DIV32(a,b) (((signed int)(a))/((signed int)(b)))