forcing tail/sibling call optimization

[Joe Buck]

Fergus Henderson writes:

> I didn't spot anything on gcc.gnu.org saying what procedure is used to
> resolve these kinds of disagreements.  I guess eventually, if/when I
> get around to implementing this, it will come down to a vote as to
> whether or not the patch gets incorporated?

Right now, Mark's under the gun to get 3.0 out, and is trying to limit
the job.  I expect that the SC will back him up on this (I certainly
will).

After 3.0 is out, we can then open the door to talking about extensions.
I suspect that a very clean patch that implements the desired
functionality will help your case a lot.

I've personally written many code generators that produce C, so certainly
my personal inclination is to support features that can make this task
easier.  But the developers now have a lot on their plate to get 3.0 out.
I suggest postponing this discussion until 3.0 has branched off.