[Cfp-interest] Part 1-3 drafts posted: Part 1

[Jim Thomas]

On Mar 20, 2013, at 1:32 PM, "Fred J. Tydeman" <tydeman at tybor.com> wrote:

> On Mon, 4 Mar 2013 19:11:50 -0800 Jim Thomas wrote:
>> 
>> http://wiki.edg.com/twiki/pub/CFP/WebHome/n1676.pdf
> 
> Some comments on part 1.
> 
> PDF page 5, lines 28-38 have a larger font than the rest of the page.

Right.

> 
> PDF page 6 lines 1-3 are the "same" as PDF page 5 lines 6-8.

Ugh. This seems to happen sometimes when I generate a document using Word's "Compare Documents …" in order to get change bars. My main draft of Part 1 doesn't have this problem.

> 
> PDF pages 10 and 11: Sections 3 and 4 are done twice.

Same problem as above.

> 
> PDF page 26, lines 26-28: Do we want SNANF32, SNANF64, SNANF80 (or
> SNANF64X), SNANF128?

No. For example, SNANF and Part 3's SNANF32 have different types.

> 
> PDF page 37, line 16, PDF page 38, line 26; both have the wrong usage
> of canonicalize().  See PDF page 46:  two arguments.

Right. Usages didn't get updated when we changed the prototypes.

> 
> PDF page 43, lines 22-32.  Seems like we need the address of x, not
> the value of x, so as to not trigger an SNaN to issignaling().  Or,
> add words saying that an SNaN is not triggered, perhaps by the use of
> a non-normal calling convention.

A classification macro is not allowed to raise any exceptions. This is stated for issignaling(), but we might need to specify it explicitly for classification macros in general.

14.7 might should be titled "Classification macros", to match C11 instead of IEC 60559.

Looking at this I noticed some references to 7.12.14 that should be 7.12.13, e.g. PDF Page 34, Line 18.

-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