Simon Riggs <simon(at)2ndQuadrant(dot)com> writes:
> Seems like GIST should be able to fake a backwards scan if needed.
Well, it tries --- there is logic in there that pays attention to the
scan direction. Like I say, I don't know the extent of the difficulty.
If we were to decide that it's unfixable, the right thing would be to
have the planner stick a materialize node on top, rather than invent
a specialized kluge for GIST.
regards, tom lane