Re: problems with IEEE NaN and Inf on Alpha

[John Eaton]

Octave doesn't fully support IEEE math on the Alpha because gcc doesn't
fully support it.  I hope this will eventually change but I have no
idea when this might happen -- as far as I can tell, it doesn't appear
to have been done for the upcoming 2.7.x release.

For what it's worth, given the proper flags, DEC's C compiler does
seem to properly handle Inf and NaN and operations on them and to
generate them, so it seems that it should be possible for gcc to do
the same.

Another option might be to fix Octave so that it is possible to
compile it with DEC C/C++, but I have no plans to do this work -- I
don't have access to DEC C++, and I am already very busy with other
things.

jwe