From: | Tomas Vondra <tomas(dot)vondra(at)postgresql(dot)org> |
---|---|
To: | pgsql-committers(at)lists(dot)postgresql(dot)org |
Subject: | pgsql: Stabilize incremental_sort tests |
Date: | 2020-04-08 16:42:15 |
Message-ID: | E1jMDmJ-0000HM-8w@gemulon.postgresql.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-committers |
Stabilize incremental_sort tests
The test never did ANALYZE on the test table, so the plans depended on
various defaults (e.g. number of groups being 200). This worked most of
the time, but with CLOBBER_CACHE_ALWAYS the autoanalyze often managed
to build accurate stats, changing the plan.
Fixed by increasing the size of test tables a bit, making the Sort a bit
more expensive than Incremental Sort. The tests were constructed to test
transitions in the Incremental Sort algorithm, and this change does not
break that.
Reviewed-by: James Coleman
Discussion: https://postgr.es/m/CAPpHfds1waRZ=NOmueYq0sx1ZSCnt+5QJvizT8ndT2=etZEeAQ@mail.gmail.com
Branch
------
master
Details
-------
https://git.postgresql.org/pg/commitdiff/cea09246e57821b8a97a6483a7df6c7345b055ef
Modified Files
--------------
src/test/regress/expected/incremental_sort.out | 32 +++++++++++++++-----------
src/test/regress/sql/incremental_sort.sql | 15 ++++++++----
2 files changed, 29 insertions(+), 18 deletions(-)
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Eisentraut | 2020-04-08 17:30:17 | pgsql: createuser: Change a fprintf to pg_log_error |
Previous Message | Tom Lane | 2020-04-08 15:23:59 | pgsql: Fix pg_dump/pg_restore to restore event trigger comments later. |