Last change
on this file since 444 was
444,
checked in by satin@…, 6 years ago
|
add newlib,libalmos-mkh, restructure shared_syscalls.h and mini-libc
|
File size:
925 bytes
|
Rev | Line | |
---|
[444] | 1 | /* |
---|
| 2 | * infinity () returns the representation of infinity. |
---|
| 3 | * Added by Cygnus Support. |
---|
| 4 | */ |
---|
| 5 | |
---|
| 6 | /* |
---|
| 7 | FUNCTION |
---|
| 8 | <<infinity>>, <<infinityf>>---representation of infinity |
---|
| 9 | |
---|
| 10 | INDEX |
---|
| 11 | infinity |
---|
| 12 | INDEX |
---|
| 13 | infinityf |
---|
| 14 | |
---|
| 15 | SYNOPSIS |
---|
| 16 | #include <math.h> |
---|
| 17 | double infinity(void); |
---|
| 18 | float infinityf(void); |
---|
| 19 | |
---|
| 20 | DESCRIPTION |
---|
| 21 | <<infinity>> and <<infinityf>> return the special number IEEE |
---|
| 22 | infinity in double- and single-precision arithmetic |
---|
| 23 | respectively. |
---|
| 24 | |
---|
| 25 | PORTABILITY |
---|
| 26 | <<infinity>> and <<infinityf>> are neither standard C nor POSIX. C and |
---|
| 27 | POSIX require macros HUGE_VAL and HUGE_VALF to be defined in math.h, which |
---|
| 28 | Newlib defines to be infinities corresponding to these archaic infinity() |
---|
| 29 | and infinityf() functions in floating-point implementations which do have |
---|
| 30 | infinities. |
---|
| 31 | |
---|
| 32 | QUICKREF |
---|
| 33 | infinity - pure |
---|
| 34 | |
---|
| 35 | */ |
---|
| 36 | |
---|
| 37 | #include "fdlibm.h" |
---|
| 38 | |
---|
| 39 | #ifndef _DOUBLE_IS_32BITS |
---|
| 40 | |
---|
| 41 | double infinity() |
---|
| 42 | { |
---|
| 43 | double x; |
---|
| 44 | |
---|
| 45 | INSERT_WORDS(x,0x7ff00000,0); |
---|
| 46 | return x; |
---|
| 47 | } |
---|
| 48 | |
---|
| 49 | #endif /* _DOUBLE_IS_32BITS */ |
---|
Note: See
TracBrowser
for help on using the repository browser.