Enable 128bit soft-fp for ia32 and ia64?

H. J. Lu hjl@lucon.org
Fri Mar 26 22:15:00 GMT 2004


Gcc 3.4 and above support __float128 for both ia32 and ia64. But there
are no supporting 128bit soft-fp. I'd like to enable soft-fp in glibc
for ia32 and ia64 so that __float128 can be used in gcc 3.4 under
ia32 and ia64. The first issue is "long double" used in soft-fp. I'd
like to change it to foat128_type, quad_float or quad_double, which
will be typedefed to "long double" by default.


H.J.



More information about the Libc-alpha mailing list