Draft "Unsafe fp optimizations" project description.

[Toon Moene]

dewar@gnat.com wrote:

> Of course, in practice the distribution of floating-point numbers seen
> in real programs is nothing like uniform.

Well, I wasn't ready to take on this problem, because it could even turn
an unbiased error into a biased one ...

> In addition, in many architectures, and in particular on the x86, you
> are using extra precision and range for intermediate computations, and
> in this case, the number of overflow cases is drastically reduced.

Good point - I didn't consider that possibility.

-- 
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)