mirror of
https://github.com/libretro/RetroArch
synced 2025-02-01 00:32:46 +00:00
148 lines
3.8 KiB
C
148 lines
3.8 KiB
C
/*
|
|
The MIT License (MIT)
|
|
|
|
Copyright (c) 2015 Lachlan Tychsen-Smith (lachlan.ts@gmail.com)
|
|
|
|
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
of this software and associated documentation files (the "Software"), to deal
|
|
in the Software without restriction, including without limitation the rights
|
|
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
copies of the Software, and to permit persons to whom the Software is
|
|
furnished to do so, subject to the following conditions:
|
|
|
|
The above copyright notice and this permission notice shall be included in
|
|
all copies or substantial portions of the Software.
|
|
|
|
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
|
THE SOFTWARE.
|
|
*/
|
|
/*
|
|
Test func : sqrtf(x)
|
|
Test Range: 0 < x < 1,000,000,000
|
|
Peak Error: ~0.0010%
|
|
RMS Error: ~0.0005%
|
|
*/
|
|
|
|
#include "math.h"
|
|
#include "math_neon.h"
|
|
|
|
void sqrtfv_c(float *x, int n, float *r)
|
|
{
|
|
|
|
float x0, x1;
|
|
float b0, b1, c0, c1;
|
|
int m0, m1;
|
|
union {
|
|
float f;
|
|
int i;
|
|
} a0, a1;
|
|
|
|
|
|
if (n & 0x1){
|
|
*r++ = sqrtf_c(*x++);
|
|
n--;
|
|
}
|
|
|
|
while(n > 0){
|
|
|
|
x0 = *x++;
|
|
x1 = *x++;
|
|
|
|
//fast invsqrt approx
|
|
a0.f = x0;
|
|
a1.f = x1;
|
|
a0.i = 0x5F3759DF - (a0.i >> 1); //VRSQRTE
|
|
a1.i = 0x5F3759DF - (a1.i >> 1); //VRSQRTE
|
|
c0 = x0 * a0.f;
|
|
c1 = x1 * a1.f;
|
|
b0 = (3.0f - c0 * a0.f) * 0.5; //VRSQRTS
|
|
b1 = (3.0f - c1 * a1.f) * 0.5; //VRSQRTS
|
|
a0.f = a0.f * b0;
|
|
a1.f = a1.f * b1;
|
|
c0 = x0 * a0.f;
|
|
c1 = x1 * a1.f;
|
|
b0 = (3.0f - c0 * a0.f) * 0.5; //VRSQRTS
|
|
b1 = (3.0f - c1 * a1.f) * 0.5; //VRSQRTS
|
|
a0.f = a0.f * b0;
|
|
a1.f = a1.f * b1;
|
|
|
|
//fast inverse approx
|
|
c0 = a0.f;
|
|
c0 = a1.f;
|
|
m0 = 0x3F800000 - (a0.i & 0x7F800000);
|
|
m1 = 0x3F800000 - (a1.i & 0x7F800000);
|
|
a0.i = a0.i + m0;
|
|
a1.i = a1.i + m1;
|
|
a0.f = 1.41176471f - 0.47058824f * a0.f;
|
|
a1.f = 1.41176471f - 0.47058824f * a1.f;
|
|
a0.i = a0.i + m0;
|
|
a1.i = a1.i + m1;
|
|
b0 = 2.0 - a0.f * c0;
|
|
b1 = 2.0 - a1.f * c1;
|
|
a0.f = a0.f * b0;
|
|
a1.f = a1.f * b1;
|
|
b0 = 2.0 - a0.f * c0;
|
|
b1 = 2.0 - a1.f * c1;
|
|
a0.f = a0.f * b0;
|
|
a1.f = a1.f * b1;
|
|
|
|
*r++ = a0.f;
|
|
*r++ = a1.f;
|
|
n -= 2;
|
|
|
|
}
|
|
}
|
|
|
|
void sqrtfv_neon(float *x, int n, float *r)
|
|
{
|
|
#if 0
|
|
asm volatile (
|
|
|
|
"tst r1, #1 \n\t" //r1 & 1
|
|
"beq 1f \n\t" //
|
|
|
|
"vld1.32 d0[0], [r0]! \n\t" //s0 = *x++
|
|
"mov ip, lr \n\t" //ip = lr
|
|
//"bl sqrtf_neon_hfp \n\t" //sqrtf_neon
|
|
"mov lr, ip \n\t" //lr = ip
|
|
"vst1.32 d0[0], [r2]! \n\t" //*r++ = r0
|
|
"subs r1, r1, #1 \n\t" //r1 = r1 - 1;
|
|
"bxeq lr \n\t" //
|
|
|
|
"1: \n\t" //
|
|
|
|
"vld1.32 d0, [r0]! \n\t" //d0 = (*x[0], *x[1]), x+=2;
|
|
|
|
//fast invsqrt approx
|
|
"vmov.f32 d1, d0 \n\t" //d1 = d0
|
|
"vrsqrte.f32 d0, d0 \n\t" //d0 = ~ 1.0 / sqrt(d0)
|
|
"vmul.f32 d2, d0, d1 \n\t" //d3 = d0 * d2
|
|
"vrsqrts.f32 d3, d2, d0 \n\t" //d4 = (3 - d0 * d3) / 2
|
|
"vmul.f32 d0, d0, d3 \n\t" //d0 = d0 * d4
|
|
"vmul.f32 d2, d0, d1 \n\t" //d3 = d0 * d2
|
|
"vrsqrts.f32 d3, d2, d0 \n\t" //d4 = (3 - d0 * d3) / 2
|
|
"vmul.f32 d0, d0, d3 \n\t" //d0 = d0 * d4
|
|
|
|
//fast reciporical approximation
|
|
"vrecpe.f32 d1, d0 \n\t" //d1 = ~ 1 / d0;
|
|
"vrecps.f32 d2, d1, d0 \n\t" //d2 = 2.0 - d1 * d0;
|
|
"vmul.f32 d1, d1, d2 \n\t" //d1 = d1 * d2;
|
|
"vrecps.f32 d2, d1, d0 \n\t" //d2 = 2.0 - d1 * d0;
|
|
"vmul.f32 d0, d1, d2 \n\t" //d0 = d1 * d2;
|
|
|
|
"vst1.64 d0, [r2]! \n\t" //*r++ = d0;
|
|
"subs r1, r1, #2 \n\t" //n = n - 2; update flags
|
|
"bgt 1b \n\t" //
|
|
|
|
::: "d0", "d1", "d2", "d3"
|
|
);
|
|
#else
|
|
sqrtfv_c(x, n, r);
|
|
#endif
|
|
}
|