Tested compiler PR patches

[Alexandre Petit-Bianco]

Bryce McKinlay writes:

> All of these PRs seem to be solved and there were no obvious
> problems or regressions caused by the patches.

Thanks a lot!

> I think they are good to be checked in, although I might have a
> closer look at 108 tommorrow...

I need to re-work on 108 to incorporate the fix described here:

  http://gcc.gnu.org/ml/gcc-patches/1999-12/msg00539.html

I'll run my own set of non regression and if the results aren't too
bad, then your new interface dispatch scheme will be good to go,
eventually.

./A