pgsql: Properly check index mark/restore in ExecSupportsMarkRestore.

From: Andrew Gierth <rhodiumtoad(at)postgresql(dot)org>
To: pgsql-committers(at)lists(dot)postgresql(dot)org
Subject: pgsql: Properly check index mark/restore in ExecSupportsMarkRestore.
Date: 2020-11-24 22:03:08
Message-ID: E1khgOy-0005ao-GC@gemulon.postgresql.org
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-committers

Properly check index mark/restore in ExecSupportsMarkRestore.

Previously this code assumed that all IndexScan nodes supported
mark/restore, which is not true since it depends on optional index AM
support functions. This could lead to errors about missing support
functions in rare edge cases of mergejoins with no sort keys, where an
unordered non-btree index scan was placed on the inner path without a
protecting Materialize node. (Normally, the fact that merge join
requires ordered input would avoid this error.)

Backpatch all the way since this bug is ancient.

Per report from Eugen Konkov on irc.

Discussion: https://postgr.es/m/87o8jn50be.fsf@news-spur.riddles.org.uk

Branch
------
REL_13_STABLE

Details
-------
https://git.postgresql.org/pg/commitdiff/6dda057043df4a56683e8fa31a74737b63da4c47

Modified Files
--------------
src/backend/executor/execAmi.c | 5 +++++
src/backend/optimizer/util/plancat.c | 2 ++
src/include/nodes/pathnodes.h | 1 +
3 files changed, 8 insertions(+)

Browse pgsql-committers by date

  From Date Subject
Next Message Michael Paquier 2020-11-25 03:19:14 pgsql: Remove catalog function currtid()
Previous Message David Rowley 2020-11-24 21:53:13 pgsql: Tidy up definitions of pg_attribute_hot and pg_attribute_cold