[cfarm-users] cfarm26 FP inaccuracies

Bruno Haible bruno at clisp.org
Sun Aug 30 21:21:39 CEST 2026


Andreas Schwab wrote:
> > making it impossible to do computations with 80-bit numbers. Thus, on musl:
> >   - 'long double' is the same as 'double', not longer,
> 
> That is not true.

Indeed: on cfarm27 (Alpine Linux), the point is not the fpucw. It's that
gcc is configured to use "--with-arch=pentium-m --with-fpmath=sse" by default.

Bruno





More information about the cfarm-users mailing list