On Sat, Apr 1, 2017 at 6:00 AM, Robert Haas <robertmh...@gmail.com> wrote: > On Fri, Mar 31, 2017 at 6:09 PM, Ashutosh Sharma <ashu.coe...@gmail.com> > wrote: >> Well, That is another option but the main idea was to be inline with >> the btree code. > > That's not a bad goal in principal, but _bt_killitems() doesn't have > any similar argument.
It was there but later got removed with some patch (may be the patch for reducing pinning and buffer content lock for btree scans). The following commit has this changes. commit 09cb5c0e7d6fbc9dee26dc429e4fc0f2a88e5272 Author: Tom Lane <t...@sss.pgh.pa.us> Date: Sun May 7 01:21:30 2006 +0000 > > (Also, speaking of consistency, why did we end up with > _hash_kill_items, with an underscore between kill and items, and > _bt_killitems, without one?) That is just to follow the naming convention in hash.h Please check the review comments for this at - . > >> Moreover, I am not sure if acquiring lwlock inside >> hashendscan (basically the place where we are trying to close down the >> things) would look good. > > Well, if that's not a good thing to do, hiding it inside some other > function doesn't make it better. okay, agreed. I will submit the patch very shortly.  - https://www.postgresql.org/message-id/cf6abd8a-77b5-f1c7-8e50-5bef461e0522%40redhat.com -- Sent via pgsql-hackers mailing list (email@example.com) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers