List of important PRs for 3.3.1, week -3.5

[Zack Weinberg]

Andrew Pinski <pinskia@physics.uc.edu> writes:

> (quoting Mark):
>  Note that Zack recently checked in better support for alternate entry
>  points.  What we really want to do is use alternate entry points; that
>  will save even more space that your approach, and be faster as well.

At the RTL level alternate entry points exist and work - sort of.  We
don't generate prologues for them, and I don't know what happens if
you put one before the FUNCTION_BEGIN note.  Not having prologues is
actually what the C++ front end wants for most of its thunks, however,
it definitely does want to put some of them before the FUNCTION_BEGIN
note... Furthermore, I got thoroughly stuck trying to figure out how
to represent alternate entry points at the tree level.

I'd be happy to work with people to get this finished.

zw