[Cfp-interest 2578] [ NB Comments ] CFP-related NB Comments (Editorial and Technical)

[JeanHeyd Meneide]

Dear Jim,

     I am sending a list of the comments I believe CFP might be
interested in, including editorial ones down below. This should
hopefully give you and CFP a chance to give input on them (I left out
all the ones related to simple formatting fixes, but ones that
influence e.g. rounding modes are left in this list).

     Let me know if you need any more information from me.

Thanks,
JeanHeyd


Editorial Comments
==============

— GB-007: single/double precision usage
— GB-147: Mentioning Annexes F, G and H with K for macros that change
header contents
— GB-151: use CMPLX over multiplication with imaginary
— GB-153: similar to GB-151
— US39-155: FE_DOWNWARD
— US42-169: nexttoward needs a "Returns" paragraph?
— GB-268: invalid frexp implementation (This is an editorial fix so
long as CFP agrees it is, despite saying technical).
— US71-275: remove "Appendix" before "IEC 60559" in §F.10.8.1 (copysign)
— GB-286: wsctofN missing like strtofN, add footnote similar to §7.31.4.1.1.
— US74-303: "_EPSILON" in §H.3 should be normalized.
— GB-322: Has identifiers for "double", missing list of identifiers
for float/long double


Technical Comments
================

— GB-157: FENV_ACCESS and execution flow with dynamic rounding modes
— GB-164: floating errors and errno
— US68-270: powr and missing (+1, infinity) argument value case handling
— GB-279, US26-075: initializers for objects of static storage
duration with "constexpr" storage class specifier.
— GB-287: strtodN, hexadecimal floating point, and
implementation-defined behavior of the base standard
— GB-288: hexadecimal floating point