/usr/include/simd.h is in soundmodem 0.18-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 | /*****************************************************************************/
/*
* simd.h -- SIMD filter procedures (aka MMX, VIS, etc).
*
* Copyright (C) 1999-2000 Thomas Sailer (sailer@ife.ee.ethz.ch)
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*
* Please note that the GPL allows you to use the driver, NOT the radio.
* In order to use the radio, you need a license from the communications
* authority of your country.
*
*/
/*****************************************************************************/
#ifndef _SIMD_H
#define _SIMD_H
/* ---------------------------------------------------------------------- */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "modem.h"
/* ---------------------------------------------------------------------- */
/*
* SIMD stuff
*/
#if !defined(__i386__)
#undef USEMMX
#endif
#if !defined(__sparc__) && !defined(__sparc64__)
#undef USEVIS
#endif
#if !defined(USEMMX) && !defined(USEVIS)
extern inline void initsimd(int enable)
{
}
extern inline int checksimd(void)
{
return 0;
}
extern inline int simdfir16(const int16_t *p1, const int16_t *p2, unsigned int nr)
{
int s = 0;
for (; nr > 0; nr--, p1++, p2++)
s += (*p1) * (*p2);
return s;
}
extern inline void simdpreparefpu(void)
{
}
#else
extern unsigned int simd_enabled;
extern void initsimd(int enable);
extern inline int checksimd(void)
{
return simd_enabled;
}
#if defined(USEMMX)
#define MMXCLOBBER "st", "st(1)", "st(2)", "st(3)", "st(4)", "st(5)", "st(6)", "st(7)"
extern inline int simdfir16(const int16_t *p1, const int16_t *p2, unsigned int nr)
{
unsigned int i, j;
int s = 0;
if (checksimd()) {
j = nr >> 2;
nr &= 3;
asm volatile("pxor\t%%mm0,%%mm0" : : : MMXCLOBBER);
for (i = 0; i < j; i++, p1 += 4, p2 += 4)
asm volatile("\n\tmovq\t%0,%%mm1"
"\n\tpmaddwd\t%1,%%mm1"
"\n\tpaddd\t%%mm1,%%mm0" : : "m" (*p1), "m" (*p2) : MMXCLOBBER);
asm volatile("\n\tmovq\t%%mm0,%%mm1 "
"\n\tpunpckhdq\t%%mm1,%%mm1"
"\n\tpaddd\t%%mm1,%%mm0"
"\n\tmovd\t%%mm0,%0" : "=m" (s) : : MMXCLOBBER);
}
for (; nr > 0; nr--, p1++, p2++)
s += (*p1) * (*p2);
return s;
}
extern inline void simdpreparefpu(void)
{
if (checksimd())
asm volatile("emms");
}
#elif defined(USEVIS)
/*extern inline*/static int simdfir16(const int16_t *p1, const int16_t *p2, unsigned int nr)
{
double dsum1, dsum2, dsum3, dsum4, arg1, arg2, arg3, arg4;
float sum, sum1, sum2;
unsigned int i, j;
int s = 0, sx[1];
if (checksimd()) {
j = nr >> 1;
nr &= 1;
asm("fzeros %0" : "=f" (sum));
for (i = 0; i < j; i++) {
asm volatile("ldda [%2] 0xd2, %0 ! ASI_FL16_P \n\t"
"ldda [%3] 0xd2, %1 ! ASI_FL16_P \n\t": "=e" (arg1), "=e" (arg2) : "r" (p1), "r" (p2));
p1++; p2++;
asm volatile("ldda [%2] 0xd2, %0 ! ASI_FL16_P \n\t"
"ldda [%3] 0xd2, %1 ! ASI_FL16_P \n\t": "=e" (arg3), "=e" (arg4) : "r" (p1), "r" (p2));
p1++; p2++;
asm volatile("fmuld8sux16 %L2,%L3,%0 \n\t"
"fmuld8ulx16 %L2,%L3,%1 \n\t" : "=&e" (dsum1), "=&e" (dsum2) : "e" (arg1), "e" (arg2));
asm volatile("fmuld8sux16 %L2,%L3,%0 \n\t"
"fmuld8ulx16 %L2,%L3,%1 \n\t" : "=&e" (dsum3), "=&e" (dsum4) : "e" (arg3), "e" (arg4));
asm volatile("fpadd32s %L1,%L2,%0" : "=f" (sum1) : "e" (dsum1), "e" (dsum2));
asm volatile("fpadd32s %L1,%L2,%0" : "=f" (sum2) : "e" (dsum3), "e" (dsum4));
asm volatile("fpadd32s %1,%2,%0" : "=f" (sum) : "f" (sum1), "f" (sum));
asm volatile("fpadd32s %1,%2,%0" : "=f" (sum) : "f" (sum2), "f" (sum));
};
asm("st %1,%0" : "=m" (sx) : "f" (sum));
s = sx[0];
}
for (; nr > 0; nr--, p1++, p2++)
s += (*p1) * (*p2);
return s;
}
extern inline void simdpreparefpu(void)
{
}
#endif
#endif
/* ---------------------------------------------------------------------- */
#endif /* _SIMD_H */
|