netcdf-c/libsrc/t_ncxx.m4
2017-03-11 13:03:17 -06:00

961 lines
19 KiB
Plaintext

dnl This is m4 source.
dnl Process using m4 to produce 'C' language file.
dnl
dnl If you see this line, you can ignore the next one.
/* Do not edit this file. It is produced from the corresponding .m4 source */
dnl
/*
* Copyright 1996, University Corporation for Atmospheric Research
* See netcdf/COPYRIGHT file for copying and redistribution conditions.
*/
/* "$Id: t_ncxx.m4,v 2.17 2005/05/20 14:48:14 russ Exp $" */
/*
* This program tests the aggregate external representation conversion
* functions "ncx_[pad_]{put,get}n_*()" declared in ncx.h.
* Unlike t_ncx.c, it only checks self consistency,
* not consistency with the xdr library.
*
* Link like this:
* cc t_ncxx.c ncx.o -o t_nxc
* (The xdr library is not needed.)
*
* If an assertion fails, there is a problem.
* Otherwise, the program is silent and has exit code 0.
*/
#if HAVE_CONFIG_H
#include <config.h>
#endif
#undef NDEBUG /* always active assert() in this file */
#include <stdio.h>
#include <string.h>
#include <limits.h>
/* alias poorly named limits.h macros */
#define SHORT_MAX SHRT_MAX
#define SHORT_MIN SHRT_MIN
#define USHORT_MAX USHRT_MAX
#include <assert.h>
#include "ncx.h"
#define X_SIZEOF_SCHAR X_SIZEOF_CHAR
#define X_LONG_MAX X_INT_MAX
#define X_LONG_MIN X_INT_MIN
#define XBSZ 1024
char ncxb[XBSZ];
char lbuf[XBSZ];
#define ArraySize(thang) (sizeof(thang)/sizeof(thang[0]))
#define eSizeOf(thang) ((size_t)(sizeof(thang[0])))
/*
* Some test data
* The ideas is that ncx_putn_type_type(...., types)
* should not return NC_ERANGE.
*/
#if SCHAR_MAX == X_SCHAR_MAX && SCHAR_MIN == X_SCHAR_MIN
static schar schars[] = {
SCHAR_MIN, SCHAR_MIN +1,
-1, 0, 1,
SCHAR_MAX - 1, SCHAR_MAX
};
#else
/* The implementation and this test assume 8 bit bytes. */
#error "Not 8 bit bytes ??"
#endif
static short shorts[] = {
#if SHORT_MAX <= X_SHORT_MAX
SHORT_MIN, SHORT_MIN + 1,
# if SCHAR_MAX < X_SHORT_MAX
SCHAR_MIN - 1, SCHAR_MIN, SCHAR_MIN + 1,
# endif
-1, 0, 1,
# if SCHAR_MAX < X_SHORT_MAX
SCHAR_MAX - 1, SCHAR_MAX, SCHAR_MAX + 1,
# endif
SHORT_MAX - 1, SHORT_MAX
#else
X_SHORT_MIN, X_SHORT_MIN + 1,
# if SCHAR_MAX < X_SHORT_MAX
SCHAR_MIN - 1, SCHAR_MIN, SCHAR_MIN + 1,
# endif
-1, 0, 1,
# if SCHAR_MAX < X_SHORT_MAX
SCHAR_MAX - 1, SCHAR_MAX, SCHAR_MAX + 1,
# endif
X_SHORT_MAX - 1, X_SHORT_MAX
#endif
};
static int ints[] = {
#if INT_MAX <= X_INT_MAX
INT_MIN, INT_MIN +1,
# if SHORT_MAX < X_INT_MAX
SHORT_MIN -1, SHORT_MIN, SHORT_MIN + 1,
# endif
# if SCHAR_MAX < X_INT_MAX
SCHAR_MIN - 1, SCHAR_MIN, SCHAR_MIN + 1,
# endif
-1, 0, 1,
# if SCHAR_MAX < X_INT_MAX
SCHAR_MAX - 1, SCHAR_MAX, SCHAR_MAX + 1,
# endif
# if SHORT_MAX < X_INT_MAX
SHORT_MAX - 1, SHORT_MAX, SHORT_MAX +1,
# endif
INT_MAX - 1, INT_MAX
#else
X_INT_MIN, X_INT_MIN +1,
# if SHORT_MAX < X_INT_MAX
SHORT_MIN -1, SHORT_MIN, SHORT_MIN + 1,
# endif
# if SCHAR_MAX < X_INT_MAX
SCHAR_MIN - 1, SCHAR_MIN, SCHAR_MIN + 1,
# endif
-1, 0, 1,
# if SCHAR_MAX < X_INT_MAX
SCHAR_MAX - 1, SCHAR_MAX, SCHAR_MAX + 1,
# endif
# if SHORT_MAX < X_INT_MAX
SHORT_MAX - 1, SHORT_MAX, SHORT_MAX +1,
# endif
X_INT_MAX - 1, X_INT_MAX
#endif /* INT */
};
/* N.B. only testing longs over X_INT range for now */
static long longs[] = {
#if LONG_MAX <= X_INT_MAX
LONG_MIN, LONG_MIN +1,
# if INT_MAX < X_INT_MAX
INT_MIN -1, INT_MIN, INT_MIN + 1,
# endif
# if SHORT_MAX < X_INT_MAX
SHORT_MIN -1, SHORT_MIN, SHORT_MIN + 1,
# endif
# if SCHAR_MAX < X_INT_MAX
SCHAR_MIN - 1, SCHAR_MIN, SCHAR_MIN + 1,
# endif
-1, 0, 1,
# if SCHAR_MAX < X_INT_MAX
SCHAR_MAX - 1, SCHAR_MAX, SCHAR_MAX + 1,
# endif
# if SHORT_MAX < X_INT_MAX
SHORT_MAX - 1, SHORT_MAX, SHORT_MAX +1,
# endif
# if INT_MAX < X_INT_MAX
INT_MAX -1, INT_MAX, INT_MAX + 1,
# endif
LONG_MAX - 1, LONG_MAX
#else
X_INT_MIN, X_INT_MIN +1,
# if SHORT_MAX < X_INT_MAX
SHORT_MIN -1, SHORT_MIN, SHORT_MIN + 1,
# endif
# if SCHAR_MAX < X_INT_MAX
SCHAR_MIN - 1, SCHAR_MIN, SCHAR_MIN + 1,
# endif
-1, 0, 1,
# if SCHAR_MAX < X_INT_MAX
SCHAR_MAX - 1, SCHAR_MAX, SCHAR_MAX + 1,
# endif
# if SHORT_MAX < X_INT_MAX
SHORT_MAX - 1, SHORT_MAX, SHORT_MAX +1,
# endif
X_INT_MAX - 1, X_INT_MAX
#endif
};
static float floats[] = {
-1.E9F,
-16777215, -16777214,
-999999,
-32769, -32768, -32767,
-129, -128, 127,
-1, 0, 1,
126, 127, 128,
32766, 32767, 32768,
999999,
16777214, 16777215, /* 2^24 -1 */
1.E9F
};
static double doubles[] = {
-1.E20,
-4503599627370495., -4503599627370494.,
-999999999999999.,
-1.E9,
-16777215, -16777214,
-999999,
-32769, -32768, -32767,
-129, -128, 127,
-1, 0, 1,
126, 127, 128,
32766, 32767, 32768,
999999,
16777214, 16777215, /* 2^24 -1 */
1.E9,
999999999999999.,
4503599627370494., 4503599627370495., /* 2^53 -1 */
1.E20
};
static uchar uchars[] = {
0, 1,
UCHAR_MAX/2 -1, UCHAR_MAX/2, UCHAR_MAX/2 +1,
UCHAR_MAX - 1, UCHAR_MAX
};
/* End of test data */
dnl dnl dnl
dnl
dnl Macros
dnl
dnl dnl dnl
dnl
dnl Upcase(str)
dnl
define(`Upcase',dnl
`dnl
translit($1, abcdefghijklmnopqrstuvwxyz, ABCDEFGHIJKLMNOPQRSTUVWXYZ)')dnl
dnl dnl dnl
dnl
dnl Xsizeof(Xtype)
dnl
define(`Xsizeof', ``X_SIZEOF_'Upcase($1)')dnl
dnl dnl dnl
define(`XMin', ``X_'Upcase($1)`_MIN'')dnl
define(`XMax', ``X_'Upcase($1)`_MAX'')dnl
dnl dnl dnl
dnl
dnl T_PUTN(XType, Type)
dnl
define(`T_PUTN',dnl
`dnl
static void
t_putn_$1_$2(char *const buf)
{
char *xp = buf;
const $2 *tp = `$2's;
size_t nelems = ArraySize(`$2's);
int status = ncx_putn_$1_$2((void **)&xp, nelems, tp);
assert(xp == buf + nelems * Xsizeof($1));
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if((double) tp[ii] > XMax($1))
{
assert(status == NC_ERANGE);
return;
}
if((double) tp[ii] < XMin($1))
{
assert(status == NC_ERANGE);
return;
}
}
assert(status == 0);
}
}
')dnl
dnl dnl dnl
dnl
dnl T_PUTN_U(XType, Type)
dnl Doesn't make signed comparisons to unsigned type
dnl
define(`T_PUTN_U',dnl
`dnl
static void
t_putn_$1_$2(char *const buf)
{
char *xp = buf;
const $2 *tp = `$2's;
size_t nelems = ArraySize(`$2's);
int status = ncx_putn_$1_$2((void **)&xp, nelems, tp);
assert(xp == buf + nelems * Xsizeof($1));
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if((double) tp[ii] > XMax($1))
{
assert(status == NC_ERANGE);
return;
}
}
assert(status == 0);
}
}
')dnl
dnl dnl dnl
dnl
dnl T_PAD_PUTN(XType, Type)
dnl
define(`T_PAD_PUTN',dnl
`dnl
static void
t_pad_putn_$1_$2(char *const buf)
{
char *xp = buf;
const $2 *tp = `$2's;
size_t nelems = ArraySize(`$2's);
const char *end = buf + nelems * Xsizeof($1);
int status = ncx_pad_putn_$1_$2((void **)&xp, nelems, tp);
assert(xp >= end);
assert((xp - end) < 4);
assert((xp - buf)%4 == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if((double) tp[ii] > XMax($1))
{
assert(status == NC_ERANGE);
return;
}
if((double) tp[ii] < XMin($1))
{
assert(status == NC_ERANGE);
return;
}
}
assert(status == 0);
}
}
')dnl
dnl dnl dnl
dnl
dnl T_GETN(XType, Type)
dnl
define(`T_GETN',dnl
`dnl
static void
t_getn_$1_$2(const char *const buf)
{
const char *xp = buf;
const $2 *tp = `$2's;
$2 *lp = ($2 *)lbuf;
size_t nelems = ArraySize(`$2's);
int status = ncx_getn_$1_$2((const void **)&xp, nelems, lp);
assert(xp == buf + nelems * Xsizeof($1));
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if(((double)tp[ii] <= XMax($1))
&& ((double)tp[ii] >= XMin($1)))
{
assert(tp[ii] == lp[ii]);
}
}
}
}
')dnl
dnl dnl dnl
dnl
dnl T_GETN_U(XType, Type)
dnl Doesn't make signed comparisons to unsigned type
dnl
define(`T_GETN_U',dnl
`dnl
static void
t_getn_$1_$2(const char *const buf)
{
const char *xp = buf;
const $2 *tp = `$2's;
$2 *lp = ($2 *)lbuf;
size_t nelems = ArraySize(`$2's);
int status = ncx_getn_$1_$2((const void **)&xp, nelems, lp);
assert(xp == buf + nelems * Xsizeof($1));
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if((double) tp[ii] <= XMax($1))
{
assert(tp[ii] == lp[ii]);
}
}
}
}
')dnl
dnl dnl dnl
dnl
dnl T_PAD_GETN(XType, Type)
dnl
define(`T_PAD_GETN',dnl
`dnl
static void
t_pad_getn_$1_$2(const char *const buf)
{
const char *xp = buf;
const $2 *tp = `$2's;
$2 *lp = ($2 *)lbuf;
size_t nelems = ArraySize(`$2's);
const char *end = buf + nelems * Xsizeof($1);
int status = ncx_pad_getn_$1_$2((const void **)&xp, nelems, lp);
assert(xp >= end);
assert((xp - end) < 4);
assert((xp - buf)%4 == 0);
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if(((double) tp[ii] <= XMax($1))
&& ((double) tp[ii] >= XMin($1)))
{
assert(tp[ii] == lp[ii]);
}
}
}
}
')dnl
dnl dnl dnl
dnl
dnl Declare & define test routines
dnl
dnl dnl dnl
T_PUTN(schar, schar)
dnl T_PUTN(schar, uchar) replaced by special case code.
dnl - we don't return conversion errors putting uchar to schar.
static void
t_putn_schar_uchar(char *const buf)
{
char *xp = buf;
const uchar *tp = uchars;
size_t nelems = ArraySize(schars);
int status = ncx_putn_schar_uchar((void **)&xp, nelems, tp);
assert(xp == buf + nelems * X_SIZEOF_SCHAR);
assert(status == 0);
}
T_PUTN(schar, short)
T_PUTN(schar, int)
T_PUTN(schar, long)
T_PUTN(schar, float)
T_PUTN(schar, double)
T_PAD_PUTN(schar, schar)
dnl T_PAD_PUTN(schar, uchar) replaced by special case code.
dnl - we don't return conversion errors putting uchar to schar.
static void
t_pad_putn_schar_uchar(char *const buf)
{
char *xp = buf;
const uchar *tp = uchars;
size_t nelems = ArraySize(uchars);
const char *end = buf + nelems * X_SIZEOF_SCHAR;
int status = ncx_pad_putn_schar_uchar((void **)&xp, nelems, tp);
assert(xp >= end);
assert((xp - end) < 4);
assert((xp - buf)%4 == 0);
assert(status == 0);
}
T_PAD_PUTN(schar, short)
T_PAD_PUTN(schar, int)
T_PAD_PUTN(schar, long)
T_PAD_PUTN(schar, float)
T_PAD_PUTN(schar, double)
T_PUTN(short, schar)
T_PUTN_U(short, uchar)
T_PUTN(short, short)
T_PUTN(short, int)
T_PUTN(short, long)
T_PUTN(short, float)
T_PUTN(short, double)
T_PAD_PUTN(short, schar)
dnl T_PAD_PUTN(short, uchar)
dnl Don't make signed comparisons to usigned type
static void
t_pad_putn_short_uchar(char *const buf)
{
char *xp = buf;
const uchar *tp = uchars;
size_t nelems = ArraySize(uchars);
const char *end = buf + nelems * X_SIZEOF_SHORT;
int status = ncx_pad_putn_short_uchar((void **)&xp, nelems, tp);
assert(xp >= end);
assert((xp - end) < 4);
assert((xp - buf)%4 == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if(tp[ii] > X_SHORT_MAX)
{
assert(status == NC_ERANGE);
return;
}
}
assert(status == 0);
}
}
T_PAD_PUTN(short, short)
T_PAD_PUTN(short, int)
T_PAD_PUTN(short, long)
T_PAD_PUTN(short, float)
T_PAD_PUTN(short, double)
T_PUTN(int, schar)
T_PUTN_U(int, uchar)
T_PUTN(int, short)
T_PUTN(int, int)
T_PUTN(int, long)
T_PUTN(int, float)
T_PUTN(int, double)
T_PUTN(float, schar)
T_PUTN_U(float, uchar)
T_PUTN(float, short)
T_PUTN(float, int)
T_PUTN(float, long)
T_PUTN(float, float)
T_PUTN(float, double)
T_PUTN(double, schar)
T_PUTN_U(double, uchar)
T_PUTN(double, short)
T_PUTN(double, int)
T_PUTN(double, long)
T_PUTN(double, float)
T_PUTN(double, double)
T_GETN(schar, schar)
dnl T_GETN(schar, uchar)
dnl - we don't return conversion errors gettin schar to uchar.
static void
t_getn_schar_uchar(const char *const buf)
{
const char *xp = buf;
const uchar *tp = uchars;
uchar *lp = (uchar *)lbuf;
size_t nelems = ArraySize(schars);
int status = ncx_getn_schar_uchar((const void **)&xp, nelems, lp);
assert(xp == buf + nelems * X_SIZEOF_SCHAR);
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
assert(tp[ii] == lp[ii]);
}
}
}
T_GETN(schar, short)
T_GETN(schar, int)
T_GETN(schar, long)
T_GETN(schar, float)
T_GETN(schar, double)
T_PAD_GETN(schar, schar)
dnl T_PAD_GETN(schar, uchar)
dnl - we don't return conversion errors gettin schar to uchar.
static void
t_pad_getn_schar_uchar(const char *const buf)
{
const char *xp = buf;
const uchar *tp = uchars;
uchar *lp = (uchar *)lbuf;
size_t nelems = ArraySize(schars);
const char *end = buf + nelems * X_SIZEOF_SCHAR;
int status = ncx_pad_getn_schar_uchar((const void **)&xp, nelems, lp);
assert(xp >= end);
assert((xp - end) < 4);
assert((xp - buf)%4 == 0);
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
assert(tp[ii] == lp[ii]);
}
}
}
T_PAD_GETN(schar, short)
T_PAD_GETN(schar, int)
T_PAD_GETN(schar, long)
T_PAD_GETN(schar, float)
T_PAD_GETN(schar, double)
T_GETN(short, schar)
T_GETN_U(short, uchar)
T_GETN(short, short)
T_GETN(short, int)
T_GETN(short, long)
T_GETN(short, float)
T_GETN(short, double)
T_PAD_GETN(short, schar)
dnl T_PAD_GETN(short, uchar)
dnl Don't make signed comparisons to usigned type
static void
t_pad_getn_short_uchar(const char *const buf)
{
const char *xp = buf;
const uchar *tp = uchars;
uchar *lp = (uchar *)lbuf;
size_t nelems = ArraySize(uchars);
const char *end = buf + nelems * X_SIZEOF_SHORT;
int status = ncx_pad_getn_short_uchar((const void **)&xp, nelems, lp);
assert(xp >= end);
assert((xp - end) < 4);
assert((xp - buf)%4 == 0);
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if((tp[ii] <= X_SHORT_MAX))
{
assert(tp[ii] == lp[ii]);
}
}
}
}
T_PAD_GETN(short, short)
T_PAD_GETN(short, int)
T_PAD_GETN(short, long)
T_PAD_GETN(short, float)
T_PAD_GETN(short, double)
T_GETN(int, schar)
T_GETN_U(int, uchar)
T_GETN(int, short)
T_GETN(int, int)
T_GETN(int, long)
T_GETN(int, float)
T_GETN(int, double)
T_GETN(float, schar)
T_GETN_U(float, uchar)
T_GETN(float, short)
dnl T_GETN(float, int)
dnl Exact conversion of int to x_float is limited by external float mantissa
static void
t_getn_float_int(const char *const buf)
{
const char *xp = buf;
const int *tp = ints;
int *lp = (int *)lbuf;
size_t nelems = ArraySize(ints);
int status = ncx_getn_float_int((const void **)&xp, nelems, lp);
assert(xp == buf + nelems * X_SIZEOF_FLOAT);
/* If the system rounds up can get NC_ERANGE */
assert(status == 0 || status == NC_ERANGE);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
/* limited by x_float mantissa nbits */
if((tp[ii] <= 16777215)
&& (tp[ii] >= -16777215))
{
assert(tp[ii] == lp[ii]);
}
}
}
}
dnl T_GETN(float, long)
dnl Exact conversion of long to x_float is limited by external float mantissa
static void
t_getn_float_long(const char *const buf)
{
const char *xp = buf;
const long *tp = longs;
long *lp = (long *)lbuf;
size_t nelems = ArraySize(longs);
int status = ncx_getn_float_long((const void **)&xp, nelems, lp);
assert(xp == buf + nelems * X_SIZEOF_FLOAT);
/* If the system rounds up can get NC_ERANGE */
assert(status == 0 || status == NC_ERANGE);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
/* limited by x_float mantissa nbits */
if((tp[ii] <= 16777215)
&& (tp[ii] >= 16777215))
{
if(tp[ii] != lp[ii])
(void) fprintf(stderr,
"%.9e != %.9e float_float (diff %.9e)\n",
(double)tp[ii], (double)lp[ii],
(double)(tp[ii] - lp[ii]));
}
}
}
}
T_GETN(float, float)
dnl T_GETN(float, double)
dnl Exact conversion of double to x_float is limited by external float mantissa
static void
t_getn_float_double(const char *const buf)
{
const char *xp = buf;
const double *tp = doubles;
double *lp = (double *)lbuf;
size_t nelems = ArraySize(doubles);
int status = ncx_getn_float_double((const void **)&xp, nelems, lp);
assert(xp == buf + nelems * X_SIZEOF_FLOAT);
assert(status == 0);
{
size_t ii;
for(ii = 0; ii < nelems; ii++)
{
if((tp[ii] <= X_FLOAT_MAX)
&& (tp[ii] >= X_FLOAT_MIN))
{
if(((float)tp[ii]) != lp[ii])
{
if(tp[ii] != 0)
{
double eps = (tp[ii] - lp[ii])/tp[ii];
if(eps > 1.19209290E-07F) /* X_FLT_EPSILON */
{
(void) fprintf(stderr,
"%.9e != %.9e float_double (eps %.9e)\n",
tp[ii], lp[ii], eps);
}
}
else
{
(void) fprintf(stderr,
"%.9e != %.9e float_double (diff %.9e)\n",
tp[ii], lp[ii], tp[ii] - lp[ii]);
}
}
}
}
}
}
T_GETN(double, schar)
T_GETN_U(double, uchar)
T_GETN(double, short)
T_GETN(double, int)
T_GETN(double, long)
T_GETN(double, float)
T_GETN(double, double)
#if defined(_CRAYIEEE) && !defined(_CRAYMPP) /* T90 */
#include <signal.h>
#endif /* T90 */
int
main(int ac, char *av[])
{
#if defined(_CRAYIEEE) && !defined(_CRAYMPP) /* T90 */
/*
* Some of the extreme test assignments in this program trigger
* floating point exceptions on CRAY T90
*/
(void) signal(SIGFPE, SIG_IGN);
#endif /* T90 */
/* x_schar */
t_putn_schar_schar(ncxb);
t_getn_schar_schar(ncxb);
t_putn_schar_uchar(ncxb);
t_getn_schar_uchar(ncxb);
t_putn_schar_short(ncxb);
t_getn_schar_short(ncxb);
t_putn_schar_int(ncxb);
t_getn_schar_int(ncxb);
t_putn_schar_long(ncxb);
t_getn_schar_long(ncxb);
t_putn_schar_float(ncxb);
t_getn_schar_float(ncxb);
t_putn_schar_double(ncxb);
t_getn_schar_double(ncxb);
/* pad x_schar */
t_pad_putn_schar_schar(ncxb);
t_getn_schar_schar(ncxb);
t_pad_getn_schar_schar(ncxb);
t_pad_putn_schar_uchar(ncxb);
t_getn_schar_uchar(ncxb);
t_pad_getn_schar_uchar(ncxb);
t_pad_putn_schar_short(ncxb);
t_getn_schar_short(ncxb);
t_pad_getn_schar_short(ncxb);
t_pad_putn_schar_int(ncxb);
t_getn_schar_int(ncxb);
t_pad_getn_schar_int(ncxb);
t_pad_putn_schar_long(ncxb);
t_getn_schar_long(ncxb);
t_pad_getn_schar_long(ncxb);
t_pad_putn_schar_float(ncxb);
t_getn_schar_float(ncxb);
t_pad_getn_schar_float(ncxb);
t_pad_putn_schar_double(ncxb);
t_getn_schar_double(ncxb);
t_pad_getn_schar_double(ncxb);
/* x_short */
t_putn_short_schar(ncxb);
t_getn_short_schar(ncxb);
t_putn_short_uchar(ncxb);
t_getn_short_uchar(ncxb);
t_putn_short_short(ncxb);
t_getn_short_short(ncxb);
t_putn_short_int(ncxb);
t_getn_short_int(ncxb);
t_putn_short_long(ncxb);
t_getn_short_long(ncxb);
t_putn_short_float(ncxb);
t_getn_short_float(ncxb);
t_putn_short_double(ncxb);
t_getn_short_double(ncxb);
/* pad x_short */
t_pad_putn_short_schar(ncxb);
t_getn_short_schar(ncxb);
t_pad_getn_short_schar(ncxb);
t_pad_putn_short_uchar(ncxb);
t_getn_short_uchar(ncxb);
t_pad_getn_short_uchar(ncxb);
t_pad_putn_short_short(ncxb);
t_getn_short_short(ncxb);
t_pad_getn_short_short(ncxb);
t_pad_putn_short_int(ncxb);
t_getn_short_int(ncxb);
t_pad_getn_short_int(ncxb);
t_pad_putn_short_long(ncxb);
t_getn_short_long(ncxb);
t_pad_getn_short_long(ncxb);
t_pad_putn_short_float(ncxb);
t_getn_short_float(ncxb);
t_pad_getn_short_float(ncxb);
t_pad_putn_short_double(ncxb);
t_getn_short_double(ncxb);
t_pad_getn_short_double(ncxb);
/* x_int */
t_putn_int_schar(ncxb);
t_getn_int_schar(ncxb);
t_putn_int_uchar(ncxb);
t_getn_int_uchar(ncxb);
t_putn_int_short(ncxb);
t_getn_int_short(ncxb);
t_putn_int_int(ncxb);
t_getn_int_int(ncxb);
t_putn_int_long(ncxb);
t_getn_int_long(ncxb);
t_putn_int_float(ncxb);
t_getn_int_float(ncxb);
t_putn_int_double(ncxb);
t_getn_int_double(ncxb);
/* x_float */
t_putn_float_schar(ncxb);
t_getn_float_schar(ncxb);
t_putn_float_uchar(ncxb);
t_getn_float_uchar(ncxb);
t_putn_float_short(ncxb);
t_getn_float_short(ncxb);
t_putn_float_int(ncxb);
t_getn_float_int(ncxb);
t_putn_float_long(ncxb);
t_getn_float_long(ncxb);
t_putn_float_float(ncxb);
t_getn_float_float(ncxb);
t_putn_float_double(ncxb);
t_getn_float_double(ncxb);
/* x_double */
t_putn_double_schar(ncxb);
t_getn_double_schar(ncxb);
t_putn_double_uchar(ncxb);
t_getn_double_uchar(ncxb);
t_putn_double_short(ncxb);
t_getn_double_short(ncxb);
t_putn_double_int(ncxb);
t_getn_double_int(ncxb);
t_putn_double_long(ncxb);
t_getn_double_long(ncxb);
t_putn_double_float(ncxb);
t_getn_double_float(ncxb);
t_putn_double_double(ncxb);
t_getn_double_double(ncxb);
return 0;
}