luke
/
ngspice
mirror of https://git.code.sf.net/p/ngspice/ngspice
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

drop ngspice internal implementation of erfc() 

which these days is guaranteed to be provided by <math.h>

note,
  our own implementation was incorrect anyway.
  it evaluated to
    erfc_ngspice(x) = erfc(fabs(x))
This commit is contained in:
rlar 2017-10-28 12:32:00 +02:00
parent ca57447f6c
commit c900cc8824
8 changed files with 1 additions and 187 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/include/ngspice/missing_math.h
View File

@ -9,10 +9,6 @@ Copyright 1999 Emmanuel Rouat
bool AlmostEqualUlps(double, double, int);
#ifndef HAVE_ERFC
extern double erfc(double);
#endif
#ifndef HAVE_LOGB
extern double logb(double);
#endif

3
src/maths/misc/Makefile.am
View File

@ -7,7 +7,6 @@ libmathmisc_la_SOURCES = \
accuracy.h \
bernoull.h \
bernoull.c \
erfc.c \
equality.c \
isinf.c \
isnan.c \
@ -18,7 +17,7 @@ libmathmisc_la_SOURCES = \
randnumb.c
EXTRA_DIST = test_accuracy.c test_erfc.c
EXTRA_DIST = test_accuracy.c
AM_CPPFLAGS = @AM_CPPFLAGS@ -I$(top_srcdir)/src/include
AM_CFLAGS = $(STATIC)

70
src/maths/misc/erfc.c
View File

@ -1,70 +0,0 @@
/**********
Copyright 1991 Regents of the University of California. All rights reserved.
Author: 1987 Kartikeya Mayaram, U. C. Berkeley CAD Group
**********/
#include "ngspice/ngspice.h"
#ifndef HAVE_ERFC
/* erfc computes the erfc(x) the code is from sedan's derfc.f */
double erfc (double x)
{
double sqrtPi, n, temp1, xSq, sum1, sum2;
sqrtPi = sqrt( M_PI );
x = ABS( x );
n = 1.0;
xSq = 2.0 * x * x;
sum1 = 0.0;
if ( x > 3.23 ) {
/* asymptotic expansion */
temp1 = exp( - x * x ) / ( sqrtPi * x );
sum2 = temp1;
while ( sum1 != sum2 ) {
sum1 = sum2;
temp1 = -1.0 * ( temp1 / xSq );
sum2 += temp1;
n += 2.0;
}
return( sum2 );
}
else {
/* series expansion for small x */
temp1 = ( 2.0 / sqrtPi ) * exp( - x * x ) * x;
sum2 = temp1;
while ( sum1 != sum2 ) {
n += 2.0;
sum1 = sum2;
temp1 *= xSq / n;
sum2 += temp1;
}
return( 1.0 - sum2 );
}
}
/* From C. Hastings, Jr., Approximations for digital computers,
Princeton Univ. Press, 1955.
Approximation accurate to within 1.5E-7
(making some assumptions about your machine's floating point mechanism)
*/
double
ierfc(double x)
{
double t, z;
t = 1/(1 + 0.3275911*x);
z = 1.061405429;
z = -1.453152027 + t * z;
z = 1.421413741 + t * z;
z = -0.284496736 + t * z;
z = 0.254829592 + t * z;
z = exp(-x*x) * t * z;
return(z);
}
#else
int Dummy_Symbol_5;
#endif

105
src/maths/misc/test_erfc.c
View File

@ -1,105 +0,0 @@
/* Paolo Nenzi 2002 - This program tests erfc function
* implementations.
*/
/*
* The situation on erfc functions in spice/cider:
*
* First we have the ierfc in spice, a sort of interpolation, which is
* fast to compute but gives is not so "good"
* Then we have derfc from cider, which is accurate but slow, the code is from sedan's derfc.f .
* Both above are only valid for x > 0.0
* Then we have glibc/os specific implementation.
*
* Proposal:
*
* Use glibc/os specific implementation as default and then test cider one.
*/
#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#ifdef HAVE_FPU_CTRL
#include <fpu_control.h>
#endif
double
derfc(double x)
{
double sqrtPi, n, temp1, xSq, sum1, sum2;
sqrtPi = sqrt( M_PI );
x = fabs( x ); /* only x > 0 interested */
n = 1.0;
xSq = 2.0 * x * x;
sum1 = 0.0;
if ( x > 3.23 ) {
/* asymptotic expansion */
temp1 = exp( - x * x ) / ( sqrtPi * x );
sum2 = temp1;
while ( sum1 != sum2 ) {
sum1 = sum2;
temp1 = -1.0 * ( temp1 / xSq );
sum2 += temp1;
n += 2.0;
}
return( sum2 );
}
else {
/* series expansion for small x */
temp1 = ( 2.0 / sqrtPi ) * exp( - x * x ) * x;
sum2 = temp1;
while ( sum1 != sum2 ) {
n += 2.0;
sum1 = sum2;
temp1 *= xSq / n;
sum2 += temp1;
}
return( 1.0 - sum2 );
}
}
double
ierfc(double x)
{
double t, z;
t = 1/(1 + 0.3275911*x);
z = 1.061405429;
z = -1.453152027 + t * z;
z = 1.421413741 + t * z;
z = -0.284496736 + t * z;
z = 0.254829592 + t * z;
z = exp(-x*x) * t * z;
return(z);
}
int main (void)
{
double x = -30.0;
double y1= 0.0, y2 = 0.0;
#ifdef HAVE_FPU_CTRL
fpu_control_t prec;
_FPU_GETCW(prec);
prec &= ~_FPU_EXTENDED;
prec |= _FPU_DOUBLE;
_FPU_SETCW(prec);
#endif
for (;(x <= 30.0);)
{
y1 = ierfc(x);
y2 = derfc(x);
printf("x: %f \t ierfc: %e \t derfc: %e \t erfc: %e\n", x, y1, y2, erfc(x) );
x = x + 1.0;
}
exit(1);
}

1
visualc/sharedspice.vcxproj
View File

@ -1101,7 +1101,6 @@
<ClCompile Include="..\src\maths\misc\accuracy.c" />
<ClCompile Include="..\src\maths\misc\bernoull.c" />
<ClCompile Include="..\src\maths\misc\equality.c" />
<ClCompile Include="..\src\maths\misc\erfc.c" />
<ClCompile Include="..\src\maths\misc\logb.c" />
<ClCompile Include="..\src\maths\misc\norm.c" />
<ClCompile Include="..\src\maths\misc\randnumb.c" />

3
visualc/src/include/ngspice/config.h
View File

@ -120,9 +120,6 @@
/* Define to 1 if you have the `endpwent' function. */
/* #undef HAVE_ENDPWENT */
/* Define to 1 if you have the `erfc' function. */
#define HAVE_ERFC 1
/* Define to 1 if you have the <fcntl.h> header file. */
#define HAVE_FCNTL_H 1

1
visualc/vngspice-fftw.vcxproj
View File

@ -1540,7 +1540,6 @@ lib /machine:x64 /def:..\..\fftw-3.3.4-dll64\libfftw3-3.def /out:$(IntDir)libfft
<ClCompile Include="..\src\maths\misc\accuracy.c" />
<ClCompile Include="..\src\maths\misc\bernoull.c" />
<ClCompile Include="..\src\maths\misc\equality.c" />
<ClCompile Include="..\src\maths\misc\erfc.c" />
<ClCompile Include="..\src\maths\misc\logb.c" />
<ClCompile Include="..\src\maths\misc\norm.c" />
<ClCompile Include="..\src\maths\misc\randnumb.c" />

1
visualc/vngspice.vcxproj
View File

@ -1506,7 +1506,6 @@
<ClCompile Include="..\src\maths\misc\accuracy.c" />
<ClCompile Include="..\src\maths\misc\bernoull.c" />
<ClCompile Include="..\src\maths\misc\equality.c" />
<ClCompile Include="..\src\maths\misc\erfc.c" />
<ClCompile Include="..\src\maths\misc\logb.c" />
<ClCompile Include="..\src\maths\misc\norm.c" />
<ClCompile Include="..\src\maths\misc\randnumb.c" />