Robert Dewar writes: > Paolo Carlini wrote: > > > Actually, sorry, __builtin_cpow returns (nan, nan) (got > > sidetracked by a strange issue I'm seeing in the C++ library), > > even "worse", so to speak... > > Well it certainly seems the right result in this case to me. Does > the standard really require the wrong result here?
F9.4.4 requires pow (x, 0) to return 1 for any x, even NaN. The controversy about what 0^0 really should be is very old, and apparently was the subject of a long flame war in the Nineteenth century... Andrew.
