[Patch] libstdc++/29989

[Paolo Carlini]

Gabriel Dos Reis wrote:

>| Right, in principle it's possible, or *was* possible, we are talking
>| about very old cruft.
>
>I seem to remember this was done because some platforms defined min
>and max as macros; I can't remember from the top of my head.  They are
>reserved names anyway, so we *can* #undef them with no harm, and I
>see no pressure in removing the #undefs.
>  
>
Likewise, I *did* not see any pressure, until people started wondering 
whether we should undef things everywhere, in <limits> too, for example. 
All in all, since we don't have *any* other undefs anywhere, I think 
it's far more consistent to take this small risk, early in Stage1. That 
said, if you feel strongly about that, then I have no problem with you 
reverting my patchlet and closing 29989 differently, simply as a 
duplicate, for example.

Paolo.