This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Kahan's Floating Point Test
- To: moshier at moshier dot ne dot mediaone dot net, mprice at atl dot lmco dot com
- Subject: Re: Kahan's Floating Point Test
- From: dewar at gnat dot com
- Date: Wed, 30 May 2001 18:34:22 -0400 (EDT)
- Cc: gcc at gcc dot gnu dot org
<<-mieee754-compliant-code-no-matter-how-slow-it-is-on-this-buggy-x86-fpu
This is not well defined. IEEE 754/854 are standards that provide for
the presence of a set of well defined floating-point operations, but the
issue of how a particular high level language *maps* into these operations
is not addressed by the standard, so you can't complain about the code
generated by a C compiler being non-IEEE compliant. Yes, you might have
some informal notions of what the mapping should be, but these notions
are not in the C standard or in the IEEE754 standard.
For more detailed discussions of this than you might want, see Sam
Figueroa's 1999 PhD thesis from NYU (Robert Dewar advisor :-)