PATCH: waitpid in natPosixProcess.cc

[Casey Marshall]

>>>>> "Tom" == Tom Tromey <tromey@redhat.com> writes:

>>>>> "Casey" == Casey Marshall <csm@gnu.org> writes:
Casey> Also: shouldn't there be a finalize method in this class, that
Casey> destroys and waits on the process?

Tom> Actually Process is explicitly defined otherwise:

Tom>     The subprocess is not killed when there are no more
Tom> references to the Process object, but rather the subprocess
Tom> continues executing asynchronously.

Tom> See also http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11801 This
Tom> has some suggestions for what to do; nothing very pleasant
Tom> though.

Ick. I think that the best course of action, then, is to keep
complaining fiercely when people use Runtime.exec.

-- 
Casey Marshall || csm@gnu.org