Java GC descriptor vs. tree_int_cst on 64-bit platforms

[Tom Tromey]

>>>>> "Hans" == Boehm, Hans <hans_boehm@hp.com> writes:

Hans> There are clearly too many different versions of this garbage
Hans> collector around at the moment.  I would personally like to move
Hans> to a model where the version in the gcc tree is the one and only
Hans> "official" one, though seriously experimental versions may still
Hans> apear elsewhere first.  That's not yet possible for several
Hans> reasons (incompatible build processes; before 6.0 there were too
Hans> many essential compile-time options; 6.0 isn't quite ready, and
Hans> hasn't yet made it into the gcc trunk.)  But all of those are
Hans> fixable, and on my list of things to fix.

Wow, that's great news.

It's always possible to do experiments, even major ones, on a branch.
I'm sure that the gcc maintainers wouldn't have a problem with
occasional branches of gcc/boehm-gc/.

Tom