[Cfp-interest 3172] Re: Range Errors in terms of underflow/overflow

Damian McGuckin damianm at esi.com.au
Sat Aug 10 01:20:09 PDT 2024


The issue is a bit more complicated.

In Section 7, no mention (from my perusal) is made of the IEEE 754 
exceptions in the math functions when range errors are mentioned.

But in Annex F, there are a few case where it is mentioned when a function 
raises an exception. But it is certainly no common. That seems somewhat 
inconsistent. By cross referencing Section 7, one can deduce the exception 
that would have been from the description of the range error. A bit messy.
Flicking back and forth is never fun and prone to being misread.

I guess I was asking whether documenting the exceptions is worth the 
effort. It may be. Or it may not be. It is not clear cut.

Note that Annex G does document ALL the exceptions (as far as I can tell).

- Damian


More information about the Cfp-interest mailing list