Could we agree to disagree and come together on improvementsandclean-up?

[Patrick Pelissier]

>> and the problem that GCC depends on GMP.
>
> Aaah ... I'm guessing this applies to gcc-4.x.x, but not gcc-3.x.x.
> Are you saying that, if you're using gcc-4.x.x (on any OS), there's not much
> gain from building a non-generic build of gmp ? (I note that the gcc version
> provided by the MinGW64 project is a 4.x.x build.)
>
> I'm not intending to be antagonistic or prevocative, btw ... it's really
> just that I don't fully understand the point that you are making. (And I
> accept that this lack of understanding is probably *my* failing, not yours.)

 If I have understood the point of the MinGW64 project, it is that GCC 4.3.x
 requires GMP and MPFR (See http://gcc.gnu.org/gcc-4.3/changes.html )
to build GCC:
 GCC uses GMP for constant folding.
 So a static generic version of GMP is enough for GCC needs.
 That's my point, and that's all I am talking about.
 It has nothing to do with the performance of any GMP on any OS.

 It seems to me that we don't talk about the same thing at all..

--
 Patrick