On Fri, Nov 2, 2018 at 7:29 PM Robert Haas <robertmh...@gmail.com> wrote: > > On Fri, Nov 2, 2018 at 7:23 AM Amit Kapila <amit.kapil...@gmail.com> wrote: > > > That doesn't seem like an unreasonable argument. I'm not sure whether > > > the right threshold is 4 or something a little bigger, but I bet it's > > > not very large. It seems important to me that before anybody thinks > > > about committing this, we construct some kind of destruction case > > > where repeated scans of the whole table are triggered as frequently as > > > possible, and then run that test with varying thresholds. > > > > Why do you think repeated scans will be a destruction case when there > > is no FSM for a small table? > > That's not what I'm saying. If we don't have the FSM, we have to > check every page of the table. If there's a workload where that > happens a lot on a table that is just under the size threshold for > creating the FSM, then it's likely to be a worst case for this patch. >
That makes sense and this is the first thing I was also worried about after looking at the initial patch and suggested a test [1] which can hit the worst case. [1] - https://www.postgresql.org/message-id/CAA4eK1%2BhP-jGYWi25-1QMedxeM_0H01s%3D%3D4-t74oEgL2EDVicw%40mail.gmail.com -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
