On 18 November 2012 08:52, Simon Riggs <simon@2ndquadrant.com> wrote:
> The difference here is that we still need to check visibility of each
> tuple, but that can be a very cheap check and never involves clog, nor
> does it dirty the page. Tuple access is reasonably expensive in
> comparison with a clog-less check on tuple xmin against snapshot xmin,
> so the extra work is negligible.
The attached *partial* patch shows how the tuple checks would work.
This should fit in neatly with the vismap skip code you've got already.
Looks to me possible to skip the all-vis hint completely, as you
suggest, plus avoid repeated checks of VM or clog.
--
Simon Riggs http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services