This is the mail archive of the
mailing list for the Java project.
Re: Precise GC (was Re: cannot build libjava/gnu/gcj/xlib/natClip.cc)
Per Bothner wrote:
> > Method arguments are not explicitly registered by this scheme
> > either. (And they could live in registers, for that matter.)
> Note that the method argument x is copied to the p structure. After
> that only p.x is used, never x itself.
Before the "p.x = x" assignment, GC is unsafe, which fits with your explanation
of safe points. Now I understand. But this imposes significant constraints on
the thread system.
Conservative, non-copying GC sure seems like a good overall compromise. In
theory it can go awry, but we'll sooner fail from network outages I'd think.