Gargage collected top-level bindings?

[Per Bothner]

Dean Ferreyra wrote:
> Which is the expected result?  In the places in our code where we use 
> fluid-let, we expect the goodbye/goodbye behavior.  Using 
> "define-variable" to define VAR gets me that result, but has this 
> UnboundLocationException issue in the presence of unrelated threads.

Hm.  ModA is treating VAR as "unknown", even though it is imported
from ModB.  I think is because of the fluid-let.
I suspect fixing this may may tie into fixing Savannah bug #11859:
https://savannah.gnu.org/bugs/?func=detailitem&item_id=11859
-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/