[MLton] share bug

[Stephen Weeks]

> I don't see an obvious connection between marking intergenerational
> pointers and the Bind exception, so I'm worried that there is
> another bug lurking around.  But, I'm willing to be convinced that
> the share fix explains the behavior.

When intergenerational pointers aren't marked correctly, there may be
bogus pointers as soon as a GC happens.  Once you have bogus pointers,
anything can happen.  I don't think what you saw is good evidence of
another bug.