> > As described, you've made
> > that problem worse because you're trying to say we don't know which
of
> > the chain entries is pointed at.
>
> There should be a flag, say HOT_CHAIN_ENTRY for the tuple the
it's called HEAP_UPDATE_ROOT
> index(es) point at. And this should be the preferred CTID for
> inserting new versions once the old one is dead.
This is not possible, see my reply to Bruce (maybe unless the whole hot
chain is dead).
(because that would need a back pointer, so readers arriving at the root
find the visible tuple)
Andreas