PostgreSQL version comparison

Ensure that parallel GIN index builds update the table's pg_class.reltuples value correctly (Jan Nidzwetzki, Tomas Vondra) §

A parallel worker could report an uninitialized value for the number of rows it processed, leading to a bogus value for reltuples, even Infinity or NaN. Such values could lead to subsequent autovacuum and autoanalyze operations never deciding that the table needs to be processed. If so, the situation will not self-heal. A manual ANALYZE command, or creation of another index, will be needed to reset reltuples to the correct value. If you have any tables with GIN indexes, it's recommended to check to see if their reltuples entries look sane. A query such as this may be helpful:

SELECT DISTINCT t.oid::regclass, t.reltuples
FROM pg_class t
  JOIN pg_index i ON t.oid = i.indrelid
  JOIN pg_class ic ON i.indexrelid = ic.oid
WHERE t.relhasindex AND ic.relam = 2742;

Correctly update foreign-data-wrapper state in a ModifyTable plan node after pruning result relations (Ayush Tiwari, Rafia Sabih) § §

Previously, if run-time partition pruning determined that some partitions of a partitioned target table need not be scanned and the table had any foreign-table partitions, a crash or erroneous behavior was likely.

Fix missed concurrent update in UPDATE with RETURNING OLD on a table that has a BEFORE UPDATE trigger (Dean Rasheed) §

If the target row was concurrently updated, then at isolation level READ COMMITTED any OLD values in RETURNING should reflect the updated row. But stale values were returned if there was a trigger (although the trigger itself, and the final output row, saw the correct values).

Fix hash join performance issue when there are multiple join keys and many NULL values (David Rowley) §

Null-keyed tuples should not get inserted into the hash table, since they will never match any other tuples. The code got this wrong if the null was in a non-last join column, bloating the hash table quite a lot if many inputs contain nulls.

Fix parsing of parenthesized OLD/NEW in RETURNING expressions (Marko Grujic) §

Expressions such as (old).colname and (old).* were mis-handled, effectively converting them to NEW references.

Clean up PlaceHolderVars more thoroughly during join removal (Richard Guo, Arne Roland) § §

This fix corrects various edge cases that could trip assertions or result in incorrect plans.

Avoid pushing WHERE clauses down past a grouping step that has a different equivalence rule (Richard Guo) §

A test on a grouping column that is grouped by a nondeterministic collation is safe to push down only if it is a comparison using that same collation. Otherwise it might filter some rows the grouping would have merged.

Avoid “cache lookup failed for collation 0” error when planner looks up statistics for a column of type "char" (Feng Wu) §

Prevent setting NO INHERIT on partitioned NOT NULL constraints via ALTER CONSTRAINT (Andreas Karlsson) §

NOT NULL constraints on partitioned tables are supposed to be inherited by all partitions, and therefore must not be marked NO INHERIT. This rule was correctly enforced by constraint creation, but not by ALTER TABLE ... ALTER CONSTRAINT.

Fix LIKE matching with nondeterministic collations and backslashes (Nitin Motiani, Tom Lane) § §

When using a nondeterministic collation, LIKE mishandled an escaped backslash (\\), treating it as effectively not there. It also did the wrong thing with a leading backslash preceding an ordinary character; in that case the backslash should be effectively ignored, but it caused the ordinary character to be matched exactly rather than allowing the nondeterministic collation to decide if there's a match.

Fix LIKE/regex optimization for indexscan with exact-match pattern (Jelte Fennema-Nio) §

Refactoring for LIKE with non-deterministic collations accidentally broke the optimization for converting a LIKE or regex exact-match pattern to an equality index condition when the index collation doesn't match the expression collation. Among other things, that made psql's \d tablename command much slower.

Fix matching of localized month/day names in to_date() (Heikki Linnakangas) § §

The matching logic misbehaved in cases where case-folding changes the byte length of the string.

Correct case-folding rules for Greek final sigma (Jeff Davis) §

If the string is preceded only by Case Ignorable characters, don't consider it to be a final sigma. This only affects the built-in pg_unicode_fast locale.

Reject out-of-range timestamp shift values in uuidv7() (Baji Shaik) §

The shift value must not be so large as to produce a timestamp out of the range that a v7 UUID can represent. Previously, a garbage UUID value was produced.

Prevent “no empty local buffer available” errors during temporary table access (Melanie Plageman) §

Limit the number of local buffers that the read streaming mechanism is allowed to use. Previously, a large value of effective_io_concurrency could allow a single stream to use all the buffers, resulting in failure.

Restore full use of shared buffer pool in VACUUM's wraparound failsafe mode (Melanie Plageman) § §

An ordinary VACUUM is limited to use just a few shared buffers, so as not to impinge too much on other processing. However, in failsafe mode we want to reclaim transaction IDs as quickly as possible, so that limit is supposed to be abandoned to allow vacuuming to proceed as fast as possible. This behavior was accidentally broken during refactoring in v18; restore it.

Fix JIT-compiled tuple deconstruction code to account correctly for virtual generated columns (David Rowley) §

Prevent accepting hot-standby connections before WAL replay has reached a consistent database state (Nikhil Sontakke) §

Avoid recording incorrect I/O operation statistics after a failed read or write (Bertrand Drouvot) §

Improve libpq's handling of out-of-memory conditions (Anthonin Bonnefoy) §

Fix libpq's trace facility to print new-style BackendKeyData and CancelRequest messages correctly (Anthonin Bonnefoy) §

Fix null-pointer crash in ecpg compiler (Jehan-Guillaume de Rorthais) §

ecpg failed on a DECLARE section containing a union nested inside a struct.

Fix issues with deferred errors in pipeline mode in psql (Michael Paquier) §

psql could get stuck or suffer an assertion failure in some scenarios where the server reports an error in response to a Sync message, such as a deferred constraint violation.

Enforce the intended upper limit for psql's special variable WATCH_INTERVAL (Sven Klemm, Daniel Gustafsson) §

If a too-large value was given, psql reported an error but applied the setting anyway.

Fix inconsistent behavior of pg_restore with --statistics or --statistics-only (Chao Li, Michael Paquier) § §

When combined with other selective-restore options such as --schema, these options failed to restore the expected items, unlike pg_dump with similar options.

Fix vacuumdb --missing-stats-only to ignore partitioned expression indexes (Baji Shaik) §

Previously, vacuumdb would always attempt to ANALYZE the partitioned table, accomplishing nothing since statistics are never created for partitioned indexes, only for their leaf indexes.

In contrib/amcheck, fix failure to report corruption of a btree metapage's allequalimage flag (Chao Li) §

In contrib/amcheck, fix query-lifespan memory leak while verifying a GIN index (Kirill Reshke) §

In contrib/btree_gist, fix sorting of bit/varbit entries during GiST index construction (Tom Lane) §

Values of bit types were sorted as though they were byteas, which did not cause any obvious failure but would result in an inefficient index, since the types' representations are different. It is recommended to reindex btree_gist indexes on bit columns after installing this update.

In contrib/dblink and contrib/postgres_fdw, ensure that a user-mapping setting for use_scram_passthrough overrides one for a foreign server (Matheus Alcantara) § §

Previously the precedence went the other way, but that is inconsistent with the behavior of other foreign-table options.

Reject setting use_scram_passthrough on contrib/dblink foreign-data wrappers (Matheus Alcantara) §

This option is only meaningful on foreign servers and user mappings, but dblink incorrectly allowed it at the FDW level as well (and then ignored it).

Fix out-of-bounds access in contrib/pg_prewarm's autoprewarm worker (Matheus Alcantara) §

The code tried to fetch a value from one past the end of an array, risking a segfault.

Detect faulty input when restoring attribute MCV statistics (Michael Paquier) §

The statistics restore functions were insufficiently careful about validating most-common-value statistics, and would accept values that could crash the planner later on.

The PostgreSQL Project thanks Jeroen Gui for reporting this problem. (CVE-2026-6575)

Improve planner's matching of partition key columns to sub-query outputs (Richard Guo) §

Strip no-op PlaceHolderVars from operands before comparing them to partition keys. This change enables partition pruning to succeed in some cases where it previously failed to recognize that a partition need not be scanned.

Fix self-join removal to handle join clauses that are bare boolean columns, e.g. ON t1.boolcol (Andrei Lepikhov, Tender Wang, Alexander Korotkov) §

Previously such a case led to a “no relation entry for relid N” error.

Fix UPDATE/DELETE ... WHERE CURRENT OF to work on tables with virtual generated columns (Satyanarayana Narlapuram, Dean Rasheed) §

Fix expansion of virtual generated columns in EXCLUDED column references in INSERT ... ON CONFLICT (Satyanarayana Narlapuram, Dean Rasheed) §

Fix spurious “indexes on virtual generated columns are not supported” errors (Robert Haas) §

Creation of an expression index could sometimes incorrectly report this error.

Fix ALTER TABLE ... SET NOT NULL to invoke object-access hook functions only after completing the catalog change (Artur Zakirov) §

Fix loss of deferrability of foreign-key triggers (Yasuo Honda) §

Previously, a foreign key defined as DEFERRABLE INITIALLY DEFERRED would behave as NOT DEFERRABLE after being set to NOT ENFORCED status and then back to ENFORCED.

If you have a foreign key with this problem, it can be repaired (after installing this update) by again setting it to NOT ENFORCED and then back to ENFORCED.

Fix WITHOUT OVERLAPS to allow domains (Jian He) §

UNIQUE/PRIMARY KEY ... WITHOUT OVERLAPS requires the no-overlap column to be a range or multirange, but it should allow a domain over such a type too.

Make the pg_aios system view's pid column show NULL not 0 when an entry has no owning process (ChangAo Chen) §

Fix under-allocation of shared memory used for parallel btree index scans (Siddharth Kothari) §

In edge cases this could result in a server crash.

Fix array overrun when too many EXPLAIN extension options are installed (Joel Jacobson) §

Correctly display join alias Vars that are used in GROUP BY (Tom Lane) §

In views containing queries like SELECT ... t1 LEFT JOIN t2 USING (x) GROUP BY x, the GROUP BY clause could be displayed incorrectly by deparsing, leading to dump/restore failures. Failures occurred only if t1.x and t2.x were not of identical data types and t1.x was the side that required an implicit coercion.

Harden tar-file parsing logic against archives it can't handle (Tom Lane) §

The tar-file reading code used in pg_basebackup and pg_verifybackup failed to verify that the input is a tar file at all, let alone that it fits into the subset of valid tar files that we can handle. This isn't a problem for the normal scenario where the input file was generated by PostgreSQL code, but it can be an issue if the input has been generated by some other tar program.

In pg_dump, preserve NO INHERIT property of NOT NULL constraints (Jian He) §

Some cases missed printing the NO INHERIT clause.

Fix mis-structured output from contrib/pg_overexplain's RANGE_TABLE option (Satyanarayana Narlapuram) §

Some fields were misplaced in JSON, YAML, and XML formats, resulting in structurally invalid output.

Fix oversight in the fix for CVE-2026-2007 (Zsolt Parragi) §

If the “bounds” array needed to be expanded, because the input contained more trigrams than the initial guess, generate_trgm_only didn't return the modified array pointer to its caller. That would lead to incorrect output from strict_word_similarity() and related functions, or in rare cases a crash. The faulty code is reached if the input string becomes longer when it's converted to lower case. The only known instances of that occur when an ICU locale is used with certain single-byte encodings.

Fix the volatility marking of json_strip_nulls() and jsonb_strip_nulls() (Andrew Dunstan) §

These functions have always been considered immutable, but refactoring in version 18 accidentally marked them stable instead. That prevents their use in index expressions and could cause unnecessary repeat evaluations in queries. This fix corrects the marking in newly-initialized database clusters (including clusters that are pg_upgrade'd to 18.3 or later). However it will not help existing clusters made using 18.0 through 18.2.

If this mistake affects your usage of these functions, the recommended fix for an existing cluster is a manual catalog update. As superuser, perform

UPDATE pg_catalog.pg_proc SET provolatile = 'i' WHERE oid IN ('3261','3262');

in each affected database. Update template0 and template1 as well, so that databases made in future will have the fix.

Avoid name collisions between user-written constraints and automatically-named NOT NULL constraints (Laurenz Albe) §

As of version 18, NOT NULL constraints have full-fledged pg_constraint entries, and therefore require names. The logic for choosing a name for an unnamed NOT NULL constraint failed to avoid conflicts with user-written constraints elsewhere in the same CREATE TABLE statement.

Harden contrib/pg_trgm against changes in string lowercasing behavior (Heikki Linnakangas) § §

Fix potential buffer overruns arising from the fact that in some locales lower-casing a string can produce more characters (not bytes) than were in the original. That behavior is new in version 18, and so is the bug.

The PostgreSQL Project thanks Heikki Linnakangas for reporting this problem. (CVE-2026-2007)

Fix inconsistent case-insensitive matching in contrib/ltree (Jeff Davis) § §

Index-related routines in ltree used a different implementation of case-folding than the primary operators did. Their behavior was equivalent only if the default collation provider was libc and the encoding was single-byte.

To fix, change the code to use case-folding with the database's default collation. This change will require reindexing indexes on ltree columns (regardless of the index access method) unless the database uses libc as collation provider and its encoding is single-byte. Without that, searches of such indexes will fail to locate relevant entries.

When using ALTER TABLE ... ADD CONSTRAINT to add a not-null constraint with an explicit name, if the column is already marked NOT NULL, require that the provided name match the existing constraint name (Álvaro Herrera, Srinath Reddy Sadipiralla) §

Fix incorrect pruning of rowmarks belonging to non-relation rangetable entries, such as subqueries (Dean Rasheed) §

This led to incorrect results if a proposed row update needed to be modified by EvalPlanQual rechecking, as could happen if there was a concurrent update to that row.

Fix expression evaluation bug for a sub-select within an array subscript (Andres Freund) §

Fix text substring search for non-deterministic collations (Laurenz Albe) §

When using a non-deterministic collation, we failed to detect a match occurring at the very end of the searched string.

Fix planner error with set-returning functions and grouping sets (Richard Guo) §

When constructing a ProjectSet plan node, the planner failed to detect that subexpressions involving grouping expressions were already computed by the input plan. This led to inefficient plans or errors such as “variable not found in subplan target list”.

Avoid incorrect optimization when a subquery's grouping clause contains a volatile or set-returning function (Richard Guo) §

The planner was willing to push down outer-query restrictions referencing such a grouping column, leading to incorrect behavior due to multiple evaluation of a volatile function, or errors caused by introduction of a set-returning function into the subquery's WHERE/HAVING clauses.

Look through PlaceHolderVar nodes when searching for statistics about an expression (Richard Guo) §

This change allows the planner to find relevant statistics about expressions pulled up from subqueries or used in GROUP BY, avoiding falling back to a default estimate. (Arguably we should adjust any found statistics to account for an increased probability of the value being NULL, but we've never done the equivalent thing for plain Vars either.) While this restriction is old, changes in PostgreSQL version 18 made PlaceHolderVars more common than before, so make the change to avoid plan regressions in affected cases.

Look through no-op PlaceHolderVar nodes when matching expressions to indexes (Richard Guo) §

Because PostgreSQL version 18 uses PlaceHolderVars in more cases than before, some queries that formerly could use an index failed to do so. Add logic to prevent that regression.

Fix planner's conversion of OR clauses to ScalarArrayOp index conditions (Tender Wang, Tom Lane) §

The code did not handle RelabelType nodes correctly, and could generate invalid expressions or fail to perform a valid conversion.

Use the correct ordering function in parallel GIN index builds (Tomas Vondra) §

The parallel code used the default ordering operator (which is determined by the column data type's btree opclass), whereas it should use the ordering function specified by the GIN opclass, if any. This led to a failure if the data type has no btree opclass, or to an invalid index if the opclass specifies an ordering function that doesn't agree with the btree opclass.

Consider grouping expressions when computing a query ID hash (Jian He) §

Previously, two queries that were the same except in GROUP BY expressions would be merged by contrib/pg_stat_statements and other users of query IDs.

Fix race condition in async I/O code (Andres Freund) §

It was possible for the result code of an asynchronous I/O operation to be overwritten before it was fetched.

Avoid failure of NUMA status views when a page has been swapped out (Tomas Vondra) §

Avoid “operation not permitted” errors when querying NUMA page status with older libnuma versions (Tomas Vondra) §

Fix aarch64-specific code to build with old (RHEL7-era) system header files (Tom Lane) § §

Fix incorrect configure probe for io_uring_queue_init_mem() (Masahiko Sawada) §

This error resulted in failure to optimize async I/O buffer allocations in autotools-based builds, though the code did work when building with meson. The main impact of the omission was slower-than-necessary backend process exits.

Fix psql's tab completion for VACUUM option values (Yugo Nagata) §

In psql command prompts, do not show a value for %P (pipeline status) when there is no server connection (Chao Li) §

This makes %P act like other prompt escape sequences whose values depend on the active connection.

Fix pg_dump's logic for collecting sequence values (Nathan Bossart) § §

pg_dump failed if a sequence was dropped concurrently with the dump, even if the sequence was not among the database objects to be dumped. Also, if the calling user lacks privileges to read a sequence's value, pg_dump emitted incorrect values rather than failing as expected.

Fix potentially-incorrect quoting of oauth_validator_libraries values by pg_dump (ChangAo Chen) §

pg_dump applied the wrong quoting rule if it needed to dump a value of this setting.

Fix excessive memory allocation in contrib/pg_buffercache (David Geier) §

The code allocated twice as much memory as it needed for NUMA page status.

Avoid crash in contrib/pg_stat_statements when an IN list contains both constants and non-constant expressions (Sami Imseih) §

Avoid incorrect optimization of variable-free HAVING clauses with grouping sets (Richard Guo) § §

Do not use parallelism in hash right semi joins (Richard Guo) §

The case does not work reliably due to a race condition in updating the join's shared hash table.

Avoid possible division-by-zero when creating ordered-append plans (Richard Guo) §

This mistake could result in incorrect selection of the cheapest path, or in an assertion failure in debug builds.

Fix planner failure with index types that can do ordered access but not index-only scans (Maxime Schoemans) §

This oversight resulted in errors like “no data returned for index-only scan”. The case does not arise with any in-core index type, but some extensions encountered the problem.

Avoid possible out-of-memory or “invalid memory alloc request size” failures during parallel GIN index build (Tomas Vondra) §

Fix rare crash when processing hashed GROUPING SETS queries (David Rowley) §

Repair faulty hash-table-size-choosing logic in hash joins (Tomas Vondra) §

Hash joins sometimes used more memory than intended, or failed to divide it in an efficient way.

Improve relation lookup logic in statistics manipulation functions (Nathan Bossart) § §

Fix pg_restore_relation_stats(), pg_clear_relation_stats(), pg_restore_attribute_stats(), and pg_clear_attribute_stats() to check privileges before acquiring lock on the target relation rather than after.

Fix crash during EvalPlanQual rechecks on partitioned tables (David Rowley, Amit Langote) §

Prevent setting a column as identity if it has a not-null constraint but the constraint is marked as invalid (Jian He) §

Identity columns must be not-null, but the check for that missed this edge case.

Re-establish special case for C collation in locale setup (Jeff Davis) §

This fixes a regression in access to shared catalogs early in backend startup, before a database has been selected. It is not known to be a problem for any core PostgreSQL code, but some extensions were broken.

In contrib/pg_stat_statements, avoid crash when two or more constants are marked as having the same location in the SQL statement text (Sami Imseih, Dmitry Dolgov) §

Change initdb default to enable data checksums (Greg Sabino Mullane) §

Checksums can be disabled with the new initdb option --no-data-checksums. pg_upgrade requires matching cluster checksum settings, so this new option can be useful to upgrade non-checksum old clusters.

Change time zone abbreviation handling (Tom Lane) §

The system will now favor the current session's time zone abbreviations before checking the server variable timezone_abbreviations. Previously timezone_abbreviations was checked first.

Deprecate MD5 password authentication (Nathan Bossart) §

Support for MD5 passwords will be removed in a future major version release. CREATE ROLE and ALTER ROLE now emit deprecation warnings when setting MD5 passwords. These warnings can be disabled by setting the md5_password_warnings parameter to off.

Change VACUUM and ANALYZE to process the inheritance children of a parent (Michael Harris) §

The previous behavior can be performed by using the new ONLY option.

Prevent COPY FROM from treating \. as an end-of-file marker when reading CSV files (Daniel Vérité, Tom Lane) § §

psql will still treat \. as an end-of-file marker when reading CSV files from STDIN. Older psql clients connecting to PostgreSQL 18 servers might experience \copy problems. This release also enforces that \. must appear alone on a line.

Disallow unlogged partitioned tables (Michael Paquier) §

Previously ALTER TABLE SET [UN]LOGGED did nothing, and the creation of an unlogged partitioned table did not cause its children to be unlogged.

Execute AFTER triggers as the role that was active when trigger events were queued (Laurenz Albe) §

Previously such triggers were run as the role that was active at trigger execution time (e.g., at COMMIT). This is significant for cases where the role is changed between queue time and transaction commit.

Remove non-functional support for rule privileges in GRANT/REVOKE (Fujii Masao) §

These have been non-functional since PostgreSQL 8.2.

Remove column pg_backend_memory_contexts.parent (Melih Mutlu) §

This is no longer needed since pg_backend_memory_contexts.path was added.

Change pg_backend_memory_contexts.level and pg_log_backend_memory_contexts() to be one-based (Melih Mutlu, Atsushi Torikoshi, David Rowley, Fujii Masao) § § §

These were previously zero-based.

Change full text search to use the default collation provider of the cluster to read configuration files and dictionaries, rather than always using libc (Peter Eisentraut) §

Clusters that default to non-libc collation providers (e.g., ICU, builtin) that behave differently than libc for characters processed by LC_CTYPE could observe changes in behavior of some full-text search functions, as well as the pg_trgm extension. When upgrading such clusters using pg_upgrade, it is recommended to reindex all indexes related to full-text search and pg_trgm after the upgrade.

Automatically remove some unnecessary table self-joins (Andrey Lepikhov, Alexander Kuzmenkov, Alexander Korotkov, Alena Rybakina) §

This optimization can be disabled using server variable enable_self_join_elimination.

Convert some IN (VALUES ...) to x = ANY ... for better optimizer statistics (Alena Rybakina, Andrei Lepikhov) §

Allow transforming OR-clauses to arrays for faster index processing (Alexander Korotkov, Andrey Lepikhov) §

Speed up the processing of INTERSECT, EXCEPT, window aggregates, and view column aliases (Tom Lane, David Rowley) § § § §

Allow the keys of SELECT DISTINCT to be internally reordered to avoid sorting (Richard Guo) §

This optimization can be disabled using enable_distinct_reordering.

Ignore GROUP BY columns that are functionally dependent on other columns (Zhang Mingli, Jian He, David Rowley) §

If a GROUP BY clause includes all columns of a unique index, as well as other columns of the same table, those other columns are redundant and can be dropped from the grouping. This was already true for non-deferred primary keys.

Allow some HAVING clauses on GROUPING SETS to be pushed to WHERE clauses (Richard Guo) § § § §

This allows earlier row filtering. This release also fixes some GROUPING SETS queries that used to return incorrect results.

Improve row estimates for generate_series() using numeric and timestamp values (David Rowley, Song Jinzhou) § §

Allow the optimizer to use Right Semi Join plans (Richard Guo) §

Semi-joins are used when needing to find if there is at least one match.

Allow merge joins to use incremental sorts (Richard Guo) §

Improve the efficiency of planning queries accessing many partitions (Ashutosh Bapat, Yuya Watari, David Rowley) § §

Allow partitionwise joins in more cases, and reduce its memory usage (Richard Guo, Tom Lane, Ashutosh Bapat) § §

Improve cost estimates of partition queries (Nikita Malakhov, Andrei Lepikhov) §

Improve SQL-language function plan caching (Alexander Pyhalov, Tom Lane) § §

Improve handling of disabled optimizer features (Robert Haas) §

Allow skip scans of btree indexes (Peter Geoghegan) § §

This allows multi-column btree indexes to be used in more cases such as when there are no restrictions on the first or early indexed columns (or there are non-equality ones), and there are useful restrictions on later indexed columns.

Allow non-btree unique indexes to be used as partition keys and in materialized views (Mark Dilger) § §

The index type must still support equality.

Allow GIN indexes to be created in parallel (Tomas Vondra, Matthias van de Meent) §

Allow values to be sorted to speed range-type GiST and btree index builds (Bernd Helmle) §

Add an asynchronous I/O subsystem (Andres Freund, Thomas Munro, Nazir Bilal Yavuz, Melanie Plageman) § § § § § § § § § § §

This feature allows backends to queue multiple read requests, which allows for more efficient sequential scans, bitmap heap scans, vacuums, etc. This is enabled by server variable io_method, with server variables io_combine_limit and io_max_combine_limit added to control it. This also enables effective_io_concurrency and maintenance_io_concurrency values greater than zero for systems without fadvise() support. The new system view pg_aios shows the file handles being used for asynchronous I/O.

Improve the locking performance of queries that access many relations (Tomas Vondra) §

Improve the performance and reduce memory usage of hash joins and GROUP BY (David Rowley, Jeff Davis) § § § § §

This also improves hash set operations used by EXCEPT, and hash lookups of subplan values.

Allow normal vacuums to freeze some pages, even though they are all-visible (Melanie Plageman) § §

This reduces the overhead of later full-relation freezing. The aggressiveness of this can be controlled by server variable and per-table setting vacuum_max_eager_freeze_failure_rate. Previously vacuum never processed all-visible pages until freezing was required.

Add server variable vacuum_truncate to control file truncation during VACUUM (Nathan Bossart, Gurjeet Singh) §

A storage-level parameter with the same name and behavior already existed.

Increase server variables effective_io_concurrency's and maintenance_io_concurrency's default values to 16 (Melanie Plageman) § §

This more accurately reflects modern hardware.

Increase the logging granularity of server variable log_connections (Melanie Plageman) §

This server variable was previously only boolean, which is still supported.

Add log_connections option to report the duration of connection stages (Melanie Plageman) §

Add log_line_prefix escape %L to output the client IP address (Greg Sabino Mullane) §

Add server variable log_lock_failures to log lock acquisition failures (Yuki Seino, Fujii Masao) § §

Specifically it reports SELECT ... NOWAIT lock failures.

Modify pg_stat_all_tables and its variants to report the time spent in VACUUM, ANALYZE, and their automatic variants (Sami Imseih) §

The new columns are total_vacuum_time, total_autovacuum_time, total_analyze_time, and total_autoanalyze_time.

Add delay time reporting to VACUUM and ANALYZE (Bertrand Drouvot, Nathan Bossart) § §

This information appears in the server log, the system views pg_stat_progress_vacuum and pg_stat_progress_analyze, and the output of VACUUM and ANALYZE when in VERBOSE mode; tracking must be enabled with the server variable track_cost_delay_timing.

Add WAL, CPU, and average read statistics output to ANALYZE VERBOSE (Anthonin Bonnefoy) § §

Add full WAL buffer count to VACUUM/ANALYZE (VERBOSE) and autovacuum log output (Bertrand Drouvot) §

Add per-backend I/O statistics reporting (Bertrand Drouvot) § §

The statistics are accessed via pg_stat_get_backend_io(). Per-backend I/O statistics can be cleared via pg_stat_reset_backend_stats().

Add pg_stat_io columns to report I/O activity in bytes (Nazir Bilal Yavuz) §

The new columns are read_bytes, write_bytes, and extend_bytes. The op_bytes column, which always equaled BLCKSZ, has been removed.

Add WAL I/O activity rows to pg_stat_io (Nazir Bilal Yavuz, Bertrand Drouvot, Michael Paquier) § § §

This includes WAL receiver activity and a wait event for such writes.

Change server variable track_wal_io_timing to control tracking WAL timing in pg_stat_io instead of pg_stat_wal (Bertrand Drouvot) §

Remove read/sync columns from pg_stat_wal (Bertrand Drouvot) § §

This removes columns wal_write, wal_sync, wal_write_time, and wal_sync_time.

Add function pg_stat_get_backend_wal() to return per-backend WAL statistics (Bertrand Drouvot) §

Per-backend WAL statistics can be cleared via pg_stat_reset_backend_stats().

Add function pg_ls_summariesdir() to specifically list the contents of PGDATA/pg_wal/summaries (Yushi Ogiwara) §

Add column pg_stat_checkpointer.num_done to report the number of completed checkpoints (Anton A. Melnikov) §

Columns num_timed and num_requested count both completed and skipped checkpoints.

Add column pg_stat_checkpointer.slru_written to report SLRU buffers written (Nitin Jadhav) §

Also, modify the checkpoint server log message to report separate shared buffer and SLRU buffer values.

Add columns to pg_stat_database to report parallel worker activity (Benoit Lobréau) §

The new columns are parallel_workers_to_launch and parallel_workers_launched.

Have query id computation of constant lists consider only the first and last constants (Dmitry Dolgov, Sami Imseih) § § §

Jumbling is used by pg_stat_statements.

Adjust query id computations to group together queries using the same relation name (Michael Paquier, Sami Imseih) §

This is true even if the tables in different schemas have different column names.

Add column pg_backend_memory_contexts.type to report the type of memory context (David Rowley) §

Add column pg_backend_memory_contexts.path to show memory context parents (Melih Mutlu) §

Add function pg_get_acl() to retrieve database access control details (Joel Jacobson) § §

Add function has_largeobject_privilege() to check large object privileges (Yugo Nagata) §

Allow ALTER DEFAULT PRIVILEGES to define large object default privileges (Takatsuka Haruka, Yugo Nagata, Laurenz Albe) §

Add predefined role pg_signal_autovacuum_worker (Kirill Reshke) §

This allows sending signals to autovacuum workers.

Add support for the OAuth authentication method (Jacob Champion, Daniel Gustafsson, Thomas Munro) §

This adds an oauth authentication method to pg_hba.conf, libpq OAuth options, a server variable oauth_validator_libraries to load token validation libraries, and a configure flag --with-libcurl to add the required compile-time libraries.

Add server variable ssl_tls13_ciphers to allow specification of multiple colon-separated TLSv1.3 cipher suites (Erica Zhang, Daniel Gustafsson) §

Change server variable ssl_groups's default to include elliptic curve X25519 (Daniel Gustafsson, Jacob Champion) §

Rename server variable ssl_ecdh_curve to ssl_groups and allow multiple colon-separated ECDH curves to be specified (Erica Zhang, Daniel Gustafsson) §

The previous name still works.

Make cancel request keys 256 bits (Heikki Linnakangas, Jelte Fennema-Nio) § §

This is only possible when the server and client support wire protocol version 3.2, introduced in this release.

Add server variable autovacuum_worker_slots to specify the maximum number of background workers (Nathan Bossart) §

With this variable set, autovacuum_max_workers can be adjusted at runtime up to this maximum without a server restart.

Allow specification of the fixed number of dead tuples that will trigger an autovacuum (Nathan Bossart, Frédéric Yhuel) §

The server variable is autovacuum_vacuum_max_threshold. Percentages are still used for triggering.

Change server variable max_files_per_process to limit only files opened by a backend (Andres Freund) §

Previously files opened by the postmaster were also counted toward this limit.

Add server variable num_os_semaphores to report the required number of semaphores (Nathan Bossart) §

This is useful for operating system configuration.

Add server variable extension_control_path to specify the location of extension control files (Peter Eisentraut, Matheus Alcantara) § §

Allow inactive replication slots to be automatically invalidated using server variable idle_replication_slot_timeout (Nisha Moond, Bharath Rupireddy) §

Add server variable max_active_replication_origins to control the maximum active replication origins (Euler Taveira) §

This was previously controlled by max_replication_slots, but this new setting allows a higher origin count in cases where fewer slots are required.

Allow the values of generated columns to be logically replicated (Shubham Khanna, Vignesh C, Zhijie Hou, Shlok Kyal, Peter Smith) § § § §

If the publication specifies a column list, all specified columns, generated and non-generated, are published. Without a specified column list, publication option publish_generated_columns controls whether generated columns are published. Previously generated columns were not replicated and the subscriber had to compute the values if possible; this is particularly useful for non-PostgreSQL subscribers which lack such a capability.

Change the default CREATE SUBSCRIPTION streaming option from off to parallel (Vignesh C) §

Allow ALTER SUBSCRIPTION to change the replication slot's two-phase commit behavior (Hayato Kuroda, Ajin Cherian, Amit Kapila, Zhijie Hou) § §

Log conflicts while applying logical replication changes (Zhijie Hou, Nisha Moond) § § § § §

Also report in new columns of pg_stat_subscription_stats.

Allow generated columns to be virtual, and make them the default (Peter Eisentraut, Jian He, Richard Guo, Dean Rasheed) § § §

Virtual generated columns generate their values when the columns are read, not written. The write behavior can still be specified via the STORED option.

Add OLD/NEW support to RETURNING in DML queries (Dean Rasheed) §

Previously RETURNING only returned new values for INSERT and UPDATE, and old values for DELETE; MERGE would return the appropriate value for the internal query executed. This new syntax allows the RETURNING list of INSERT/UPDATE/DELETE/MERGE to explicitly return old and new values by using the special aliases old and new. These aliases can be renamed to avoid identifier conflicts.

Allow foreign tables to be created like existing local tables (Zhang Mingli) §

The syntax is CREATE FOREIGN TABLE ... LIKE.

Allow LIKE with nondeterministic collations (Peter Eisentraut) §

Allow text position search functions with nondeterministic collations (Peter Eisentraut) §

These used to generate an error.

Add builtin collation provider PG_UNICODE_FAST (Jeff Davis) §

This locale supports case mapping, but sorts in code point order, not natural language order.

Allow VACUUM and ANALYZE to process partitioned tables without processing their children (Michael Harris) §

This is enabled with the new ONLY option. This is useful since autovacuum does not process partitioned tables, just its children.

Add functions to modify per-relation and per-column optimizer statistics (Corey Huinker) § § §

The functions are pg_restore_relation_stats(), pg_restore_attribute_stats(), pg_clear_relation_stats(), and pg_clear_attribute_stats().

Add server variable file_copy_method to control the file copying method (Nazir Bilal Yavuz) §

This controls whether CREATE DATABASE ... STRATEGY=FILE_COPY and ALTER DATABASE ... SET TABLESPACE uses file copy or clone.

Allow the specification of non-overlapping PRIMARY KEY, UNIQUE, and foreign key constraints (Paul A. Jungwirth) § §

This is specified by WITHOUT OVERLAPS for PRIMARY KEY and UNIQUE, and by PERIOD for foreign keys, all applied to the last specified column.

Allow CHECK and foreign key constraints to be specified as NOT ENFORCED (Amul Sul) § §

This also adds column pg_constraint.conenforced.

Require primary/foreign key relationships to use either deterministic collations or the the same nondeterministic collations (Peter Eisentraut) §

The restore of a pg_dump, also used by pg_upgrade, will fail if these requirements are not met; schema changes must be made for these upgrade methods to succeed.

Store column NOT NULL specifications in pg_constraint (Álvaro Herrera, Bernd Helmle) § §

This allows names to be specified for NOT NULL constraint. This also adds NOT NULL constraints to foreign tables and NOT NULL inheritance control to local tables.

Allow ALTER TABLE to set the NOT VALID attribute of NOT NULL constraints (Rushabh Lathia, Jian He) §

Allow modification of the inheritability of NOT NULL constraints (Suraj Kharage, Álvaro Herrera) § §

The syntax is ALTER TABLE ... ALTER CONSTRAINT ... [NO] INHERIT.

Allow NOT VALID foreign key constraints on partitioned tables (Amul Sul) §

Allow dropping of constraints ONLY on partitioned tables (Álvaro Herrera) §

This was previously erroneously prohibited.

Add REJECT_LIMIT to control the number of invalid rows COPY FROM can ignore (Atsushi Torikoshi) §

This is available when ON_ERROR = 'ignore'.

Allow COPY TO to copy rows from populated materialized views (Jian He) §

Add COPY LOG_VERBOSITY level silent to suppress log output of ignored rows (Atsushi Torikoshi) §

This new level suppresses output for discarded input rows when on_error = 'ignore'.

Disallow COPY FREEZE on foreign tables (Nathan Bossart) §

Previously, the COPY worked but the FREEZE was ignored, so disallow this command.

Automatically include BUFFERS output in EXPLAIN ANALYZE (Guillaume Lelarge, David Rowley) §

Add full WAL buffer count to EXPLAIN (WAL) output (Bertrand Drouvot) §

In EXPLAIN ANALYZE, report the number of index lookups used per index scan node (Peter Geoghegan) §

Modify EXPLAIN to output fractional row counts (Ibrar Ahmed, Ilia Evdokimov, Robert Haas) § §

Add memory and disk usage details to Material, Window Aggregate, and common table expression nodes to EXPLAIN output (David Rowley, Tatsuo Ishii) § § § §

Add details about window function arguments to EXPLAIN output (Tom Lane) §

Add Parallel Bitmap Heap Scan worker cache statistics to EXPLAIN ANALYZE (David Geier, Heikki Linnakangas, Donghang Lin, Alena Rybakina, David Rowley) §

Indicate disabled nodes in EXPLAIN ANALYZE output (Robert Haas, David Rowley, Laurenz Albe) § § §

Improve Unicode full case mapping and conversion (Jeff Davis) § §

This adds the ability to do conditional and title case mapping, and case map single characters to multiple characters.

Allow jsonb null values to be cast to scalar types as NULL (Tom Lane) §

Previously such casts generated an error.

Add optional parameter to json{b}_strip_nulls to allow removal of null array elements (Florents Tselai) §

Add function array_sort() which sorts an array's first dimension (Junwang Zhao, Jian He) §

Add function array_reverse() which reverses an array's first dimension (Aleksander Alekseev) §

Add function reverse() to reverse bytea bytes (Aleksander Alekseev) §

Allow casting between integer types and bytea (Aleksander Alekseev) §

The integer values are stored as bytea two's complement values.

Update Unicode data to Unicode 16.0.0 (Peter Eisentraut) §

Add full text search stemming for Estonian (Tom Lane) §

Improve the XML error codes to more closely match the SQL standard (Tom Lane) §

These errors are reported via SQLSTATE.

Add function casefold() to allow for more sophisticated case-insensitive matching (Jeff Davis) §

This allows more accurate comparisons, i.e., a character can have multiple upper or lower case equivalents, or upper or lower case conversion changes the number of characters.

Allow MIN()/MAX() aggregates on arrays and composite types (Aleksander Alekseev, Marat Buharov) § §

Add a WEEK option to EXTRACT() (Tom Lane) §

Improve the output EXTRACT(QUARTER ...) for negative values (Tom Lane) §

Add roman numeral support to to_number() (Hunaid Sohail) §

This is accessed via the RN pattern.

Add UUID version 7 generation function uuidv7() (Andrey Borodin) §

This UUID value is temporally sortable. Function alias uuidv4() has been added to explicitly generate version 4 UUIDs.

Add functions crc32() and crc32c() to compute CRC values (Aleksander Alekseev) §

Add math functions gamma() and lgamma() (Dean Rasheed) §

Allow => syntax for named cursor arguments in PL/pgSQL (Pavel Stehule) §

We previously only accepted :=.

Allow regexp_match[es]()/regexp_like()/regexp_replace()/regexp_count()/regexp_instr()/regexp_substr()/regexp_split_to_table()/regexp_split_to_array() to use named arguments (Jian He) §

Add function PQfullProtocolVersion() to report the full, including minor, protocol version number (Jacob Champion, Jelte Fennema-Nio) §

Add libpq connection parameters and environment variables to specify the minimum and maximum acceptable protocol version for connections (Jelte Fennema-Nio) § §

Report search_path changes to the client (Alexander Kukushkin, Jelte Fennema-Nio, Tomas Vondra) § §

Add PQtrace() output for all message types, including authentication (Jelte Fennema-Nio) § § § § §

Add libpq connection parameter sslkeylogfile which dumps out SSL key material (Abhishek Chanda, Daniel Gustafsson) §

This is useful for debugging.

Modify some libpq function signatures to use int64_t (Thomas Munro) §

These previously used pg_int64, which is now deprecated.

Allow psql to parse, bind, and close named prepared statements (Anthonin Bonnefoy, Michael Paquier) § §

This is accomplished with new commands \parse, \bind_named, and \close_prepared.

Add psql backslash commands to allowing issuance of pipeline queries (Anthonin Bonnefoy) § § §

The new commands are \startpipeline, \syncpipeline, \sendpipeline, \endpipeline, \flushrequest, \flush, and \getresults.

Allow adding pipeline status to the psql prompt and add related state variables (Anthonin Bonnefoy) §

The new prompt character is %P and the new psql variables are PIPELINE_SYNC_COUNT, PIPELINE_COMMAND_COUNT, and PIPELINE_RESULT_COUNT.

Allow adding the connection service name to the psql prompt or access it via psql variable (Michael Banck) §

Add psql option to use expanded mode on all list commands (Dean Rasheed) §

Adding backslash suffix x enables this.

Change psql's \conninfo to use tabular format and include more information (Álvaro Herrera, Maiquel Grassi, Hunaid Sohail) §

Add function's leakproof indicator to psql's \df+, \do+, \dAo+, and \dC+ outputs (Yugo Nagata) §

Add access method details for partitioned relations in \dP+ (Justin Pryzby) §

Add default_version to the psql \dx extension output (Magnus Hagander) §

Add psql variable WATCH_INTERVAL to set the default \watch wait time (Daniel Gustafsson) §

Change initdb to default to enabling checksums (Greg Sabino Mullane) § §

The new initdb option --no-data-checksums disables checksums.

Add initdb option --no-sync-data-files to avoid syncing heap/index files (Nathan Bossart) §

initdb option --no-sync is still available to avoid syncing any files.

Add vacuumdb option --missing-stats-only to compute only missing optimizer statistics (Corey Huinker, Nathan Bossart) § §

This option can only be run by superusers and can only be used with options --analyze-only and --analyze-in-stages.

Add pg_combinebackup option -k/--link to enable hard linking (Israel Barth Rubio, Robert Haas) §

Only some files can be hard linked. This should not be used if the backups will be used independently.

Allow pg_verifybackup to verify tar-format backups (Amul Sul) §

If pg_rewind's --source-server specifies a database name, use it in --write-recovery-conf output (Masahiko Sawada) §

Add pg_resetwal option --char-signedness to change the default char signedness (Masahiko Sawada) §

Add pg_dump option --statistics (Jeff Davis) § §

Add pg_dump and pg_dumpall option --sequence-data to dump sequence data that would normally be excluded (Nathan Bossart) § §

Add pg_dump, pg_dumpall, and pg_restore options --statistics-only, --no-statistics, --no-data, and --no-schema (Corey Huinker, Jeff Davis) §

Add option --no-policies to disable row level security policy processing in pg_dump, pg_dumpall, pg_restore (Nikolay Samokhvalov) §

This is useful for migrating to systems with different policies.

Allow pg_upgrade to preserve optimizer statistics (Corey Huinker, Jeff Davis, Nathan Bossart) § § § §

Extended statistics are not preserved. Also add pg_upgrade option --no-statistics to disable statistics preservation.

Allow pg_upgrade to process database checks in parallel (Nathan Bossart) § § § § § § § § § § §

This is controlled by the existing --jobs option.

Add pg_upgrade option --swap to swap directories rather than copy, clone, or link files (Nathan Bossart) §

This mode is potentially the fastest.

Add pg_upgrade option --set-char-signedness to set the default char signedness of new cluster (Masahiko Sawada) § §

This is to handle cases where a pre-PostgreSQL 18 cluster's default CPU signedness does not match the new cluster.

Add pg_createsubscriber option --all to create logical replicas for all databases (Shubham Khanna) §

Add pg_createsubscriber option --clean to remove publications (Shubham Khanna) § §

Add pg_createsubscriber option --enable-two-phase to enable prepared transactions (Shubham Khanna) §

Add pg_recvlogical option --enable-failover to specify failover slots (Hayato Kuroda) §

Also add option --enable-two-phase as a synonym for --two-phase, and deprecate the latter.

Allow pg_recvlogical --drop-slot to work without --dbname (Hayato Kuroda) §

Separate the loading and running of injection points (Michael Paquier, Heikki Linnakangas) § §

Injection points can now be created, but not run, via INJECTION_POINT_LOAD(), and such injection points can be run via INJECTION_POINT_CACHED().

Support runtime arguments in injection points (Michael Paquier) §

Allow inline injection point test code with IS_INJECTION_POINT_ATTACHED() (Heikki Linnakangas) §

Improve the performance of processing long JSON strings using SIMD (Single Instruction Multiple Data) (David Rowley) §

Speed up CRC32C calculations using x86 AVX-512 instructions (Raghuveer Devulapalli, Paul Amonson) §

Add ARM Neon and SVE CPU intrinsics for popcount (integer bit counting) (Chiranmoy Bhattacharya, Devanga Susmitha, Rama Malladi) § §

Improve the speed of numeric multiplication and division (Joel Jacobson, Dean Rasheed) § § § §

Add configure option --with-libnuma to enable NUMA awareness (Jakub Wartak, Bertrand Drouvot) § § §

The function pg_numa_available() reports on NUMA awareness, and system views pg_shmem_allocations_numa and pg_buffercache_numa which report on shared memory distribution across NUMA nodes.

Add TOAST table to pg_index to allow for very large expression indexes (Nathan Bossart) §

Remove column pg_attribute.attcacheoff (David Rowley) §

Add column pg_class.relallfrozen (Melanie Plageman) §

Add amgettreeheight, amconsistentequality, and amconsistentordering to the index access method API (Mark Dilger) § §

Add GiST support function stratnum() (Paul A. Jungwirth) §

Record the default CPU signedness of char in pg_controldata (Masahiko Sawada) §

Add support for Python "Limited API" in PL/Python (Peter Eisentraut) § §

This helps prevent problems caused by Python 3.x version mismatches.

Change the minimum supported Python version to 3.6.8 (Jacob Champion) §

Remove support for OpenSSL versions older than 1.1.1 (Daniel Gustafsson) § §

If LLVM is enabled, require version 14 or later (Thomas Munro) §

Add macro PG_MODULE_MAGIC_EXT to allow extensions to report their name and version (Andrei Lepikhov) §

This information can be access via the new function pg_get_loaded_modules().

Document that SPI_connect()/SPI_connect_ext() always returns success (SPI_OK_CONNECT) (Stepan Neretin) §

Errors are always reported via ereport().

Add documentation section about API and ABI compatibility (David Wheeler, Peter Eisentraut) §

Remove the experimental designation of Meson builds on Windows (Aleksander Alekseev) §

Remove configure options --disable-spinlocks and --disable-atomics (Thomas Munro) § §

Thirty-two-bit atomic operations are now required.

Remove support for the HPPA/PA-RISC architecture (Tom Lane) §

Add extension pg_logicalinspect to inspect logical snapshots (Bertrand Drouvot) §

Add extension pg_overexplain which adds debug details to EXPLAIN output (Robert Haas) §

Add output columns to postgres_fdw_get_connections() (Hayato Kuroda, Sagar Dilip Shedge) § § § §

New output column used_in_xact indicates if the foreign data wrapper is being used by a current transaction, closed indicates if it is closed, user_name indicates the user name, and remote_backend_pid indicates the remote backend process identifier.

Allow SCRAM authentication from the client to be passed to postgres_fdw servers (Matheus Alcantara, Peter Eisentraut) §

This avoids storing postgres_fdw authentication information in the database, and is enabled with the postgres_fdw use_scram_passthrough connection option. libpq uses new connection parameters scram_client_key and scram_server_key.

Allow SCRAM authentication from the client to be passed to dblink servers (Matheus Alcantara) §

Add on_error and log_verbosity options to file_fdw (Atsushi Torikoshi) §

These control how file_fdw handles and reports invalid file rows.

Add reject_limit to control the number of invalid rows file_fdw can ignore (Atsushi Torikoshi) §

This is active when ON_ERROR = 'ignore'.

Add configurable variable min_password_length to passwordcheck (Emanuele Musella, Maurizio Boriani) §

This controls the minimum password length.

Have pgbench report the number of failed, retried, or skipped transactions in per-script reports (Yugo Nagata) §

Add isn server variable weak to control invalid check digit acceptance (Viktor Holmberg) §

This was previously only controlled by function isn_weak().

Allow values to be sorted to speed btree_gist index builds (Bernd Helmle, Andrey Borodin) §

Add amcheck check function gin_index_check() to verify GIN indexes (Grigory Kryachko, Heikki Linnakangas, Andrey Borodin) §

Add functions pg_buffercache_evict_relation() and pg_buffercache_evict_all() to evict unpinned shared buffers (Nazir Bilal Yavuz) §

The existing function pg_buffercache_evict() now returns the buffer flush status.

Allow extensions to install custom EXPLAIN options (Robert Haas, Sami Imseih) § § §

Allow extensions to use the server's cumulative statistics API (Michael Paquier) § §

Allow the queries of CREATE TABLE AS and DECLARE to be tracked by pg_stat_statements (Anthonin Bonnefoy) §

They are also now assigned query ids.

Allow the parameterization of SET values in pg_stat_statements (Greg Sabino Mullane, Michael Paquier) §

This reduces the bloat caused by SET statements with differing constants.

Add pg_stat_statements columns to report parallel activity (Guillaume Lelarge) §

The new columns are parallel_workers_to_launch and parallel_workers_launched.

Add pg_stat_statements.wal_buffers_full to report full WAL buffers (Bertrand Drouvot) §

Add pgcrypto algorithms sha256crypt and sha512crypt (Bernd Helmle) §

Add CFB mode to pgcrypto encryption and decryption (Umar Hayat) §

Add function fips_mode() to report the server's FIPS mode (Daniel Gustafsson) §

Add pgcrypto server variable builtin_crypto_enabled to allow disabling builtin non-FIPS mode cryptographic functions (Daniel Gustafsson, Joe Conway) §

This is useful for guaranteeing FIPS mode behavior.