Jeremy Drake <pgsql@jdrake.com> writes:
> On Mon, 5 Sep 2011, Bruce Momjian wrote:
>> What would we investigate except a compiler bug?
> To me, simply chalking it up to some uncharacterized compiler bug is still
> quite a bit of black magic.
If there were some reason to believe either that it wasn't a compiler
bug, or that there would be something reasonable we could do to work
around it, then I'd be interested in pressing further. But on the
strength of what we have now, neither of those things seem real likely.
I'm with Bruce on thinking that it's probably not going to repay further
effort.
regards, tom lane