Hi Paul, > > 'ceil' and 'ceil-posix'. > > But this isn't a POSIX thing, it's an IEC 60559 thing, right?
Not exactly. IEC 60559 is said to be the same as IEEE 754, and in the copy I have the only text about the rounding-to-integer operations is this one: "5.5. Round Floating-Point Number to Integer Value It shall be possible to round a floating-point number to an integral valued floating-point number in the same format. The rounding shall be as specified in Section 4, with the understanding that when rounding to nearest, if the difference between the unrounded operand and the rounded result is exactly one half, the rounded result is even." and neither section 4 nor section 6.3 say something about ceil(-0.0). Likewise in IEEE 854. So, the behaviour with minus zero originates in section F.9.6 of ISO C 99, not earlier. > So I don't see why the stricter modules would have the "-posix" > suffix. The name should be something like "ceil-ieee-fp", no? 'ceil-posix' is not very good, because it's an optional part of POSIX and an optional part of ISO C 99 as well. But minus zero is an IEEE 754 invention, therefore I agree the name could go in this direction. Even shorter, call it 'ceil-ieee'. The GCC option is called "-mieee" or "-mieee-conformant". > Anyway, I suspect the stricter modules would be overkill, no matter > what the name, and that nobody would use them. So I suggest > simply dropping the stricter requirements for now. If I didn't have a couple of hours of free time now, I would agree with you :-) Bruno