Possible C++ bug: construction/destruction of global objects

[Evan Lavelle]

Loren James Rittle wrote:

> If a standard says or strongly implies that exit() should be callable
> in a global constructor, we have a bug traceable back to gcc circa 2.95.

Thanks Loren. I guess there's no shortage of language lawyers here, but 
I'm also asking the question on c.l.c++; I'll let you know if there's 
any useful feedback.

Evan