[Cfp-interest] N1676 comments spreadsheet updates

[Jim Thomas]

On Apr 12, 2013, at 12:26 PM, "Fred J. Tydeman" <tydeman at tybor.com> wrote:

> On Fri, 12 Apr 2013 10:30:42 -0700 Jim Thomas wrote:
>> 
>> F.3 specifies that issignaling (and all the other 
>> classification macros), raise no floating-point exception 
>> if the argument is a variable, or any other expression 
>> whose value is represented in the format of the its 
>> semantic type, even if the value is a signaling NaN.
> 
> I think that still has a problem.  An expression
> evaluated in its semantic type could raise any FP
> exception during argument evaluation:
>  isnan( 1.L * SNANL );
>  isnan( 0.L / 0.L );
>  isinf( 1.L / 0.L );
>  isinf( LDBL_MAX * LDBL_MAX );
>  iszero( LDBL_MIN * LDBL_MIN );
>  isnormal( 1.L / 3.L );

Yes, argument evaluation can raise exceptions. It can for fabs and the negation operator too. But that's different from the operation's raising an exception.

-Jim

> 
> 
> ---
> Fred J. Tydeman        Tydeman Consulting
> tydeman at tybor.com      Testing, numerics, programming
> +1 (775) 287-5904      Vice-chair of PL22.11 (ANSI "C")
> Sample C99+FPCE tests: http://www.tybor.com
> Savers sleep well, investors eat well, spenders work forever.
> 
> _______________________________________________
> Cfp-interest mailing list
> Cfp-interest at oakapple.net
> http://mailman.oakapple.net/mailman/listinfo/cfp-interest