2012/10/2 Marc Glisse <marc.gli...@inria.fr>: > Here I am talking of a library issue: the wording that says that there are > sufficient overloads such that integer types call the double version of math > functions. It is fairly obvious that it doesn't apply to abs(long) for > instance which has an explicit overload. For short or unsigned, I still read > it as saying that it converts to double...
This really looks like a problem of the Standard Library specification to me and a corresponding issue should be submitted. In fact the wording can be interpreted that mixing <cstdlib> with <cmath> would imply two different versions of std::abs(int) because of different required return types. I will prepare a corresponding submission to the LWG. - Daniel