On Fri, Dec 16, 2011 at 18:08, Greg Smith <greg@2ndquadrant.com> wrote:
> I don't think this
> is going to reach ready to commit in the next few days though, so I'm going
> to mark it as returned for this CommitFest.
Fair enough, I just hope this doesn't get dragged into the next
commitfest without feedback.
> perhaps someone can help sorting out the parsing area
> that's accidentally being decelerated.
Well the slowdown isn't "accidental", I think it's expected since I'm
adding a fair bit of code to expression processing (which isn't all
pretty).
It could be reduced by doing the caching decisions in a 2nd pass,
inside ExecInitExpr, but it would mean adding an extra field to
'struct Expr' and require a significant rewrite of the patch. I'm not
sure if it's worthwhile to attempt that approach:
http://archives.postgresql.org/pgsql-hackers/2011-12/msg00483.php
Regards,
Marti