Re: [MIT-Scheme-devel] Cleaning weak hash tables in secondary GC daemon?

[Taylor R Campbell]

   Date: Tue, 20 Sep 2011 16:38:07 -0700
   From: Matt Birkholz <address@hidden>

   Sorry, somehow I missed that twice.  So I do not need to call
   trigger-secondary-gc-daemons!  Thanks.  I must have hit "Aborting!:
   out of memory" for other reasons.  I will have to let a GtkLayout demo
   run down to 4Kwords just to see it go.

The `out of memory' message happens essentially simultaneously with
running secondary GC daemons.  If you want to run them but you don't
want to abort, you'll have to do that yourself.

But maybe you really want to use a GC finalizer or a `primary' GC
daemon.  What are you using secondary GC daemons for?

   > Ideally, in this case, Scheme would instead expand the heap, rather
   > than aborting; [...] But [...] we don't have a scalable GC [...]

   Why is that?  Just because 26bits is tight?  58bits aught to provide
   for a bigger newspace...

What I meant by `we don't have a scalable GC' is that our pauses are
long (and unbounded above) and our throughput is low (and unbounded
below).  This doesn't matter much with a 26-bit limit on the size of
the heap, but it does matter with a 58-bit limit.