Two or three initial observations on the fp prining patch

[Paolo Carlini]

Jerry Quinn wrote:

> I suppose the optimal case would be for the compiler to make these
> definitions available to us when the installed compiler runs. In lieu
> of that, it looks like we'll need to access the tm.h file when the
> library is configured, right?

Hummm. I'm not 100% sure to get your point. I see the issue like this:
when we build the library we need some definitions in order to compile
a floatconv.cc correctly targeted.

To this effect, we can included 'tm.h' in floatconv.cc (*), since it
provides those definitions.

Paolo.

(*) In order to actually do that we need to make sure that all the files
included by tm.h are in the search path. But this is only a low level
practical issue.