On Tue, Apr 18, 2017 at 5:39 AM, Jonathan Wakely <jwakely.gcc@xxxxxxxxx> wrote: > On 18 April 2017 at 10:37, Jeffrey Walton wrote: >> Intrinsics could have worked everywhere, but something else was provided. > > Because something else was requested. Lets clear that up now. Here's the quote of what was requested: </QUOTE> Please make all headers for intrinsics be includable without special compiler flags. [Example redacted]. Not only that, please make all the intrinsics functions be defined and ready to be used. [Example redacted]. </QUOTE> > If you want a different change, > ask for it, don't complain it wasn't provided in response to a > different request. I wanted the same thing that was requested, but its a moot point now. The earliest it would be usable in the mainstream is 5 or 7 years or so. There's no guarantee it will be implemented, and no guarantee when a distro will pick it up. (The example showed freestanding functions that always returned a constant value. I'd like to use it in member functions, and functions that returned something other than a constant too.). Jeff
