On Wed, Feb 24, 2021 at 8:13 PM Justin Pryzby <pry...@telsasoft.com> wrote: > I think 8e12f4a25 wasn't quite aggressive enough in its changes, and I had > another patch laying around. I rebased and came up with this.
See my remarks/questions about vacuum_cleanup_index_scale_factor addressed to Masahiko from a little earlier. I think that it might make sense to just remove it. It might even make sense to disable it in the backbranches -- that approach might be better than trying to fix the "IndexVacuumInfo.num_heap_tuples is only representative of the heap relation at the end of the VACUUM when considered within btvacuumcleanup()" bug. (Though I'm less confident on this second point about a backpatchable fix.) -- Peter Geoghegan
