Messages in this thread | | | Subject | Re: floating-point abuse in 2.1.113 | From | Andreas Schwab <> | Date | 07 Aug 1998 11:33:31 +0200 |
| |
"H. Peter Anvin" <hpa@transmeta.com> writes:
|> > > |> > > > I don't think this is a big problem anymore with mostly all CPUs supporting |> > > > standard IEEE floating point arithmetic. Is there any Linux target that has |> > > > non-IEEE FP? I doubt it. |> > > |> > > VAX. Also, for example i386 has 80-bit floating point, which no other |> > > system have, and I think Alpha and Sparc64 have 128-bit. |> > |> > Sparc32 has 128-bit as well. But both 80-bit and 128-bit floating point are |> > IEEE as well... So a generic emulator could help. |> > |> |> Only it is arbitrary-precision, which is very slow (it may or may not |> matter, of course.)
It doesn't matter for the compiler. And by the way, gcc *does* emulate floating point by default if XFmode (80/96 bits) or TFmode (128 bits) is to be supported, in order to get precise, consistent results on all platforms, whether cross compiling or not. Look for REAL_ARITHMETIC in the gcc sources.
-- Andreas Schwab "And now for something schwab@issan.informatik.uni-dortmund.de completely different" schwab@gnu.org
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |