2008-06-30 21:53:51 -04:00
|
|
|
#ifndef __UTILS_H_
|
|
|
|
#define __UTILS_H_
|
|
|
|
|
|
|
|
/* these functions byteswap any-size units --------------------- */
|
|
|
|
void bswap(byte_t *s, size_t n);
|
|
|
|
void bswap_to(byte_t *dest, byte_t *src, size_t n);
|
|
|
|
void bswap_buffer(byte_t *data, size_t sz, size_t npts);
|
|
|
|
/* ------------------------------------------------------------- */
|
|
|
|
|
|
|
|
/* reverse the order of elements of any size, in place or out of place */
|
|
|
|
/* n is the number of elements */
|
|
|
|
void memreverse(char *a, size_t n, size_t elsz);
|
|
|
|
void memreverse_to(char *dest, char *a, size_t n, size_t elsz);
|
|
|
|
|
|
|
|
/* swap the contents of two buffers */
|
|
|
|
void memswap(char *a, char *b, size_t sz);
|
|
|
|
|
|
|
|
/* allocating aligned blocks ----------------------------------- */
|
|
|
|
void *malloc_aligned(size_t size, size_t align_size);
|
|
|
|
void free_aligned(void *ptr);
|
|
|
|
void *realloc_aligned(void *ptr, size_t size, size_t align_size);
|
|
|
|
/* ------------------------------------------------------------- */
|
|
|
|
|
|
|
|
int dbl_equals(double a, double b);
|
|
|
|
int flt_equals(float a, float b);
|
|
|
|
void dbl_tolerance(double tol);
|
|
|
|
void flt_tolerance(float tol);
|
|
|
|
int double_exponent(double d);
|
|
|
|
double double_mantissa(double d);
|
|
|
|
int float_exponent(float f);
|
|
|
|
float float_mantissa(float f);
|
|
|
|
void snprint_real(char *s, size_t cnt, double r,
|
|
|
|
int width, // printf field width, or 0
|
|
|
|
int dec, // # decimal digits desired, recommend 16
|
|
|
|
// # of zeros in .00...0x before using scientific notation
|
|
|
|
// recommend 3-4 or so
|
|
|
|
int max_digs_rt,
|
|
|
|
// # of digits left of decimal before scientific notation
|
|
|
|
// recommend 10
|
|
|
|
int max_digs_lf);
|
|
|
|
void snprint_cplx(char *s, size_t cnt, double re, double im,
|
|
|
|
// args to pass on to snprint_real
|
|
|
|
int width, int dec,
|
|
|
|
int max_digs_rt, int max_digs_lf,
|
|
|
|
// print spaces around sign in a+bi
|
|
|
|
int spflag);
|
|
|
|
|
2009-03-13 18:26:44 -04:00
|
|
|
char *uint2str(char *dest, size_t len, uint64_t num, uint32_t base);
|
2009-02-01 00:41:43 -05:00
|
|
|
int str2int(char *str, size_t len, int64_t *res, uint32_t base);
|
|
|
|
int isdigit_base(char c, int base);
|
2008-12-30 23:45:08 -05:00
|
|
|
|
2008-06-30 21:53:51 -04:00
|
|
|
extern double trunc(double x);
|
|
|
|
|
|
|
|
STATIC_INLINE double fpart(double arg)
|
|
|
|
{
|
|
|
|
return arg - trunc(arg);
|
|
|
|
}
|
|
|
|
|
|
|
|
#define ipart(x) trunc(x)
|
|
|
|
|
|
|
|
numerictype_t effective_numerictype(double r);
|
|
|
|
double conv_to_double(void *data, numerictype_t tag);
|
|
|
|
void conv_from_double(void *data, double d, numerictype_t tag);
|
|
|
|
int64_t conv_to_int64(void *data, numerictype_t tag);
|
|
|
|
uint64_t conv_to_uint64(void *data, numerictype_t tag);
|
|
|
|
int32_t conv_to_int32(void *data, numerictype_t tag);
|
|
|
|
uint32_t conv_to_uint32(void *data, numerictype_t tag);
|
|
|
|
#ifdef BITS64
|
|
|
|
#define conv_to_long conv_to_int64
|
|
|
|
#define conv_to_ulong conv_to_uint64
|
|
|
|
#else
|
|
|
|
#define conv_to_long conv_to_int32
|
|
|
|
#define conv_to_ulong conv_to_uint32
|
|
|
|
#endif
|
|
|
|
int cmp_same_lt(void *a, void *b, numerictype_t tag);
|
|
|
|
int cmp_same_eq(void *a, void *b, numerictype_t tag);
|
|
|
|
int cmp_lt(void *a, numerictype_t atag, void *b, numerictype_t btag);
|
2009-04-15 19:54:43 -04:00
|
|
|
int cmp_eq(void *a, numerictype_t atag, void *b, numerictype_t btag,
|
|
|
|
int equalnans);
|
2008-06-30 21:53:51 -04:00
|
|
|
|
|
|
|
#ifdef ARCH_X86_64
|
|
|
|
# define LEGACY_REGS "=Q"
|
|
|
|
#else
|
|
|
|
# define LEGACY_REGS "=q"
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#if !defined(__INTEL_COMPILER) && (defined(ARCH_X86) || defined(ARCH_X86_64))
|
|
|
|
STATIC_INLINE u_int16_t ByteSwap16(u_int16_t x)
|
|
|
|
{
|
|
|
|
__asm("xchgb %b0,%h0" :
|
|
|
|
LEGACY_REGS (x) :
|
|
|
|
"0" (x));
|
|
|
|
return x;
|
|
|
|
}
|
|
|
|
#define bswap_16(x) ByteSwap16(x)
|
|
|
|
|
|
|
|
STATIC_INLINE u_int32_t ByteSwap32(u_int32_t x)
|
|
|
|
{
|
|
|
|
#if __CPU__ > 386
|
|
|
|
__asm("bswap %0":
|
|
|
|
"=r" (x) :
|
|
|
|
#else
|
|
|
|
__asm("xchgb %b0,%h0\n"\
|
|
|
|
" rorl $16,%0\n"
|
|
|
|
" xchgb %b0,%h0":
|
|
|
|
LEGACY_REGS (x) :
|
|
|
|
#endif
|
|
|
|
"0" (x));
|
|
|
|
return x;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define bswap_32(x) ByteSwap32(x)
|
|
|
|
|
|
|
|
STATIC_INLINE u_int64_t ByteSwap64(u_int64_t x)
|
|
|
|
{
|
|
|
|
#ifdef ARCH_X86_64
|
|
|
|
__asm("bswap %0":
|
|
|
|
"=r" (x) :
|
|
|
|
"0" (x));
|
|
|
|
return x;
|
|
|
|
#else
|
|
|
|
register union { __extension__ u_int64_t __ll;
|
|
|
|
u_int32_t __l[2]; } __x;
|
|
|
|
asm("xchgl %0,%1":
|
|
|
|
"=r"(__x.__l[0]),"=r"(__x.__l[1]):
|
|
|
|
"0"(bswap_32((unsigned long)x)),"1"(bswap_32((unsigned long)(x>>32))));
|
|
|
|
return __x.__ll;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
#define bswap_64(x) ByteSwap64(x)
|
|
|
|
|
|
|
|
#else
|
|
|
|
|
|
|
|
#define bswap_16(x) (((x) & 0x00ff) << 8 | ((x) & 0xff00) >> 8)
|
|
|
|
|
|
|
|
#ifdef __INTEL_COMPILER
|
|
|
|
#define bswap_32(x) _bswap(x)
|
|
|
|
#else
|
|
|
|
#define bswap_32(x) \
|
|
|
|
((((x) & 0xff000000) >> 24) | (((x) & 0x00ff0000) >> 8) | \
|
|
|
|
(((x) & 0x0000ff00) << 8) | (((x) & 0x000000ff) << 24))
|
|
|
|
#endif
|
|
|
|
|
|
|
|
STATIC_INLINE u_int64_t ByteSwap64(u_int64_t x)
|
|
|
|
{
|
|
|
|
union {
|
|
|
|
u_int64_t ll;
|
|
|
|
u_int32_t l[2];
|
|
|
|
} w, r;
|
|
|
|
w.ll = x;
|
|
|
|
r.l[0] = bswap_32 (w.l[1]);
|
|
|
|
r.l[1] = bswap_32 (w.l[0]);
|
|
|
|
return r.ll;
|
|
|
|
}
|
|
|
|
#define bswap_64(x) ByteSwap64(x)
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#endif
|