g++ and aliasing bools

[Paolo Carlini]

Dan Nicolaescu wrote:

> We can relax the checks ss soon as we convinve ourselves that more
> cases are acceptable.
> Is a patch like this acceptable at this point, or should I wait until
> 3.1 branches?

Hi Dan,

it's nice that, after some many words, you are willing to actually implement
what Daniel, Joe and the others have proposed and discussed!

However, I think that you should wait a little more:

    http://gcc.gnu.org/ml/gcc/2002-01/msg01685.html

Please, post some actual testsuite and benchmarking results as soon as you have
them!

Cheers,
Paolo.