fixed the definition of moo_fltmax_t
This commit is contained in:
parent
44810da2d6
commit
5672b62e27
@ -361,15 +361,15 @@
|
||||
* The moo_fltmax_t type defines the largest floating-pointer number type
|
||||
* ever supported.
|
||||
*/
|
||||
#if MOO_SIZEOF___FLOAT128 >= MOO_SIZEOF_FLT_T
|
||||
#if MOO_SIZEOF___FLOAT128 >= MOO_SIZEOF_FLTBAS_T
|
||||
/* the size of long double may be equal to the size of __float128
|
||||
* for alignment on some platforms */
|
||||
typedef __float128 moo_fltmax_t;
|
||||
# define MOO_SIZEOF_FLTMAX_T MOO_SIZEOF___FLOAT128
|
||||
# define MOO_FLTMAX_REQUIRE_QUADMATH 1
|
||||
#else
|
||||
typedef moo_flt_t moo_fltmax_t;
|
||||
# define MOO_SIZEOF_FLTMAX_T MOO_SIZEOF_FLT_T
|
||||
typedef moo_fltbas_t moo_fltmax_t;
|
||||
# define MOO_SIZEOF_FLTMAX_T MOO_SIZEOF_FLTBAS_T
|
||||
# undef MOO_FLTMAX_REQUIRE_QUADMATH
|
||||
#endif
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user