finitel: Use correct GCC builtin
__builtin_isinf_sign doesn't reflect the correct return value for NaN to emulate finite function. Use __builtin_isfinite instead whichg is available since GCC 4.4 just as __builtin_isinf_sign. * libm/common/sl_finite.c (finitel): Use __builtin_isfinite. Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
This commit is contained in:
parent
3e446e9723
commit
373f81d17c
|
@ -19,7 +19,7 @@ finitel (long double x)
|
||||||
Some architectures for example have an 80-bit long double whereas
|
Some architectures for example have an 80-bit long double whereas
|
||||||
others use 128-bits. We use macros and comiler builtin functions
|
others use 128-bits. We use macros and comiler builtin functions
|
||||||
to avoid specific knowledge of the long double format. */
|
to avoid specific knowledge of the long double format. */
|
||||||
return __builtin_isinf_sign (x) == 0;
|
return __builtin_isfinite (x);
|
||||||
#endif
|
#endif
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
Loading…
Reference in New Issue