[Cfp-interest] Part 1: nextup(DBL_MAX)

[Jim Thomas]

On May 17, 2012, at 9:33 PM, Fred J. Tydeman wrote:

> On Thu, 17 May 2012 21:06:02 -0700 Jim Thomas wrote:
>> 
>> 
>>> nextafter(DBL_MAX,INFINITY) is a range error (main text)
>>> and an "overflow" exception (Annex F).
>>> 
>>> nextup(DBL_MAX) might be treated the same (unless we
>>> prohibit errors / exceptions).
>> 
>> 754:1985 specified an overflow for the nextafter case above.
>> But 754:2008 says nextup is quiet except for sNaN arguments.
> 
> Agreed and I know that (but that is 754).
> But our CFP document is silent on these cases being allowed or prohibited.

Does adding the statements in Annex F (per my May 17, 2012 4:43:51 PM PDT message) address your concern?

-Jim

> 
> ---
> Fred J. Tydeman        Tydeman Consulting
> tydeman at tybor.com      Testing, numerics, programming
> +1 (775) 358-9748      Vice-chair of PL22.11 (ANSI "C")
> Sample C99+FPCE tests: http://www.tybor.com
> Savers sleep well, investors eat well, spenders work forever.