[LCP]Detecting overflow of floating point calculations

[Vincent Penquerc'h]

> it conforms to BSD 4.3.  Does anyone know how to make it
> portable?

I guess if the target uses IEEE fp format, it's a simple
check of the right bit patterns.
Also, check for denormals, which are quite slow (and tell
you that an overflow is near).

-- 
Vincent Penquerc'h 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.linux.org.au/pipermail/linuxcprogramming/attachments/20020626/82c7b5f4/attachment.htm