release-9.2.sgml

<!-- doc/src/sgml/release-9.2.sgml -->
<!-- See header comment in release.sgml about typical markup -->

 <sect1 id="release-9-2-20">
  <title>Release 9.2.20</title>

  <note>
  <title>Release Date</title>
  <simpara>2017-02-09</simpara>
  </note>

  <para>
   This release contains a variety of fixes from 9.2.19.
   For information about new features in the 9.2 major release, see
   <xref linkend="release-9-2">.
  </para>

  <sect2>
   <title>Migration to Version 9.2.20</title>

   <para>
    A dump/restore is not required for those running 9.2.X.
   </para>

   <para>
    However, if your installation has been affected by the bug described in
    the first changelog entry below, then after updating you may need
    to take action to repair corrupted indexes.
   </para>

   <para>
    Also, if you are upgrading from a version earlier than 9.2.11,
    see <xref linkend="release-9-2-11">.
   </para>

  </sect2>

  <sect2>
   <title>Changes</title>

   <itemizedlist>

    <listitem>
     <para>
      Fix a race condition that could cause indexes built
      with <command>CREATE INDEX CONCURRENTLY</> to be corrupt
      (Pavan Deolasee, Tom Lane)
     </para>

     <para>
      If <command>CREATE INDEX CONCURRENTLY</> was used to build an index
      that depends on a column not previously indexed, then rows
      updated by transactions that ran concurrently with
      the <command>CREATE INDEX</> command could have received incorrect
      index entries.  If you suspect this may have happened, the most
      reliable solution is to rebuild affected indexes after installing
      this update.
     </para>
    </listitem>

    <listitem>
     <para>
      Unconditionally WAL-log creation of the <quote>init fork</> for an
      unlogged table (Michael Paquier)
     </para>

     <para>
      Previously, this was skipped when <xref linkend="guc-wal-level">
      = <literal>minimal</>, but actually it's necessary even in that case
      to ensure that the unlogged table is properly reset to empty after a
      crash.
     </para>
    </listitem>

    <listitem>
<!--
Author: Fujii Masao <fujii@postgresql.org>
Branch: REL9_2_STABLE [38bec1805] 2017-01-25 07:02:25 +0900
-->
     <para>
      Fix WAL page header validation when re-reading segments (Takayuki
      Tsunakawa, Amit Kapila)
     </para>

     <para>
      In corner cases, a spurious <quote>out-of-sequence TLI</> error
      could be reported during recovery.
     </para>
    </listitem>

    <listitem>
     <para>
      If the stats collector dies during hot standby, restart it (Takayuki
      Tsunakawa)
     </para>
    </listitem>

    <listitem>
     <para>
      Check for interrupts while hot standby is waiting for a conflicting
      query (Simon Riggs)
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid constantly respawning the autovacuum launcher in a corner case
      (Amit Khandekar)
     </para>

     <para>
      This fix avoids problems when autovacuum is nominally off and there
      are some tables that require freezing, but all such tables are
      already being processed by autovacuum workers.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix check for when an extension member object can be dropped (Tom Lane)
     </para>

     <para>
      Extension upgrade scripts should be able to drop member objects,
      but this was disallowed for serial-column sequences, and possibly
      other cases.
     </para>
    </listitem>

    <listitem>
     <para>
      Make sure <command>ALTER TABLE</> preserves index tablespace
      assignments when rebuilding indexes (Tom Lane, Michael Paquier)
     </para>

     <para>
      Previously, non-default settings
      of <xref linkend="guc-default-tablespace"> could result in broken
      indexes.
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent dropping a foreign-key constraint if there are pending
      trigger events for the referenced relation (Tom Lane)
     </para>

     <para>
      This avoids <quote>could not find trigger <replaceable>NNN</></quote>
      or <quote>relation <replaceable>NNN</> has no triggers</quote> errors.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix processing of OID column when a table with OIDs is associated to
      a parent with OIDs via <command>ALTER TABLE ... INHERIT</> (Amit
      Langote)
     </para>

     <para>
      The OID column should be treated the same as regular user columns in
      this case, but it wasn't, leading to odd behavior in later
      inheritance changes.
     </para>
    </listitem>

    <listitem>
     <para>
      Check for serializability conflicts before reporting
      constraint-violation failures (Thomas Munro)
     </para>

     <para>
      When using serializable transaction isolation, it is desirable
      that any error due to concurrent transactions should manifest
      as a serialization failure, thereby cueing the application that
      a retry might succeed.  Unfortunately, this does not reliably
      happen for duplicate-key failures caused by concurrent insertions.
      This change ensures that such an error will be reported as a
      serialization error if the application explicitly checked for
      the presence of a conflicting key (and did not find it) earlier
      in the transaction.
     </para>
    </listitem>

    <listitem>
     <para>
      Ensure that column typmods are determined accurately for
      multi-row <literal>VALUES</> constructs (Tom Lane)
     </para>

     <para>
      This fixes problems occurring when the first value in a column has a
      determinable typmod (e.g., length for a <type>varchar</> value) but
      later values don't share the same limit.
     </para>
    </listitem>

    <listitem>
     <para>
      Throw error for an unfinished Unicode surrogate pair at the end of a
      Unicode string (Tom Lane)
     </para>

     <para>
      Normally, a Unicode surrogate leading character must be followed by a
      Unicode surrogate trailing character, but the check for this was
      missed if the leading character was the last character in a Unicode
      string literal (<literal>U&amp;'...'</>) or Unicode identifier
      (<literal>U&amp;"..."</>).
     </para>
    </listitem>

    <listitem>
     <para>
      Ensure that a purely negative text search query, such
      as <literal>!foo</>, matches empty <type>tsvector</>s (Tom Dunstan)
     </para>

     <para>
      Such matches were found by GIN index searches, but not by sequential
      scans or GiST index searches.
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent crash when <function>ts_rewrite()</> replaces a non-top-level
      subtree with an empty query (Artur Zakirov)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix performance problems in <function>ts_rewrite()</> (Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <function>ts_rewrite()</>'s handling of nested NOT operators
      (Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <function>array_fill()</> to handle empty arrays properly (Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix one-byte buffer overrun in <function>quote_literal_cstr()</>
      (Heikki Linnakangas)
     </para>

     <para>
      The overrun occurred only if the input consisted entirely of single
      quotes and/or backslashes.
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent multiple calls of <function>pg_start_backup()</>
      and <function>pg_stop_backup()</> from running concurrently (Michael
      Paquier)
     </para>

     <para>
      This avoids an assertion failure, and possibly worse things, if
      someone tries to run these functions in parallel.
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid discarding <type>interval</>-to-<type>interval</> casts
      that aren't really no-ops (Tom Lane)
     </para>

     <para>
      In some cases, a cast that should result in zeroing out
      low-order <type>interval</> fields was mistakenly deemed to be a
      no-op and discarded.  An example is that casting from <type>INTERVAL
      MONTH</> to <type>INTERVAL YEAR</> failed to clear the months field.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <application>pg_dump</> to dump user-defined casts and transforms
      that use built-in functions (Stephen Frost)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix possible <application>pg_basebackup</> failure on standby
      server when including WAL files (Amit Kapila, Robert Haas)
     </para>
    </listitem>

    <listitem>
     <para>
      Ensure that the Python exception objects we create for PL/Python are
      properly reference-counted (Rafa de la Torre, Tom Lane)
     </para>

     <para>
      This avoids failures if the objects are used after a Python garbage
      collection cycle has occurred.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix PL/Tcl to support triggers on tables that have <literal>.tupno</>
      as a column name (Tom Lane)
     </para>

     <para>
      This matches the (previously undocumented) behavior of
      PL/Tcl's <command>spi_exec</> and <command>spi_execp</> commands,
      namely that a magic <literal>.tupno</> column is inserted only if
      there isn't a real column named that.
     </para>
    </listitem>

    <listitem>
     <para>
      Allow DOS-style line endings in <filename>~/.pgpass</> files,
      even on Unix (Vik Fearing)
     </para>

     <para>
      This change simplifies use of the same password file across Unix and
      Windows machines.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix one-byte buffer overrun if <application>ecpg</> is given a file
      name that ends with a dot (Takayuki Tsunakawa)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <application>psql</>'s tab completion for <command>ALTER DEFAULT
      PRIVILEGES</> (Gilles Darold, Stephen Frost)
     </para>
    </listitem>

    <listitem>
     <para>
      In <application>psql</>, treat an empty or all-blank setting of
      the <envar>PAGER</> environment variable as meaning <quote>no
      pager</> (Tom Lane)
     </para>

     <para>
      Previously, such a setting caused output intended for the pager to
      vanish entirely.
     </para>
    </listitem>

    <listitem>
     <para>
      Improve <filename>contrib/dblink</>'s reporting of
      low-level <application>libpq</> errors, such as out-of-memory
      (Joe Conway)
     </para>
    </listitem>

    <listitem>
     <para>
      On Windows, ensure that environment variable changes are propagated
      to DLLs built with debug options (Christian Ullrich)
     </para>
    </listitem>

    <listitem>
     <para>
      Sync our copy of the timezone library with IANA release tzcode2016j
      (Tom Lane)
     </para>

     <para>
      This fixes various issues, most notably that timezone data
      installation failed if the target directory didn't support hard
      links.
     </para>
    </listitem>

    <listitem>
     <para>
      Update time zone data files to <application>tzdata</> release 2016j
      for DST law changes in northern Cyprus (adding a new zone
      Asia/Famagusta), Russia (adding a new zone Europe/Saratov), Tonga,
      and Antarctica/Casey.
      Historical corrections for Italy, Kazakhstan, Malta, and Palestine.
      Switch to preferring numeric zone abbreviations for Tonga.
     </para>
    </listitem>

   </itemizedlist>

  </sect2>
 </sect1>

 <sect1 id="release-9-2-19">
  <title>Release 9.2.19</title>

  <note>
  <title>Release Date</title>
  <simpara>2016-10-27</simpara>
  </note>

  <para>
   This release contains a variety of fixes from 9.2.18.
   For information about new features in the 9.2 major release, see
   <xref linkend="release-9-2">.
  </para>

  <sect2>
   <title>Migration to Version 9.2.19</title>

   <para>
    A dump/restore is not required for those running 9.2.X.
   </para>

   <para>
    However, if you are upgrading from a version earlier than 9.2.11,
    see <xref linkend="release-9-2-11">.
   </para>

  </sect2>

  <sect2>
   <title>Changes</title>

   <itemizedlist>

    <listitem>
     <para>
      Fix EvalPlanQual rechecks involving CTE scans (Tom Lane)
     </para>

     <para>
      The recheck would always see the CTE as returning no rows, typically
      leading to failure to update rows that were recently updated.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix improper repetition of previous results from hashed aggregation in
      a subquery (Andrew Gierth)
     </para>

     <para>
      The test to see if we can reuse a previously-computed hash table of
      the aggregate state values neglected the possibility of an outer query
      reference appearing in an aggregate argument expression.  A change in
      the value of such a reference should lead to recalculating the hash
      table, but did not.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <command>EXPLAIN</> to emit valid XML when
      <xref linkend="guc-track-io-timing"> is on (Markus Winand)
     </para>

     <para>
      Previously the XML output-format option produced syntactically invalid
      tags such as <literal>&lt;I/O-Read-Time&gt;</>.  That is now
      rendered as <literal>&lt;I-O-Read-Time&gt;</>.
     </para>
    </listitem>

    <listitem>
     <para>
      Suppress printing of zeroes for unmeasured times
      in <command>EXPLAIN</> (Maksim Milyutin)
     </para>

     <para>
      Certain option combinations resulted in printing zero values for times
      that actually aren't ever measured in that combination.  Our general
      policy in <command>EXPLAIN</> is not to print such fields at all, so
      do that consistently in all cases.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix timeout length when <command>VACUUM</> is waiting for exclusive
      table lock so that it can truncate the table (Simon Riggs)
     </para>

     <para>
      The timeout was meant to be 50 milliseconds, but it was actually only
      50 microseconds, causing <command>VACUUM</> to give up on truncation
      much more easily than intended.  Set it to the intended value.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix bugs in merging inherited <literal>CHECK</> constraints while
      creating or altering a table (Tom Lane, Amit Langote)
     </para>

     <para>
      Allow identical <literal>CHECK</> constraints to be added to a parent
      and child table in either order.  Prevent merging of a valid
      constraint from the parent table with a <literal>NOT VALID</>
      constraint on the child.  Likewise, prevent merging of a <literal>NO
      INHERIT</> child constraint with an inherited constraint.
     </para>
    </listitem>

    <listitem>
     <para>
      Remove artificial restrictions on the values accepted
      by <function>numeric_in()</> and <function>numeric_recv()</>
      (Tom Lane)
     </para>

     <para>
      We allow numeric values up to the limit of the storage format (more
      than <literal>1e100000</>), so it seems fairly pointless
      that <function>numeric_in()</> rejected scientific-notation exponents
      above 1000.  Likewise, it was silly for <function>numeric_recv()</> to
      reject more than 1000 digits in an input value.
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid very-low-probability data corruption due to testing tuple
      visibility without holding buffer lock (Thomas Munro, Peter Geoghegan,
      Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix file descriptor leakage when truncating a temporary relation of
      more than 1GB (Andres Freund)
     </para>
    </listitem>

    <listitem>
     <para>
      Disallow starting a standalone backend with <literal>standby_mode</>
      turned on (Michael Paquier)
     </para>

     <para>
      This can't do anything useful, since there will be no WAL receiver
      process to fetch more WAL data; and it could result in misbehavior
      in code that wasn't designed with this situation in mind.
     </para>
    </listitem>

    <listitem>
     <para>
      Don't try to share SSL contexts across multiple connections
      in <application>libpq</> (Heikki Linnakangas)
     </para>

     <para>
      This led to assorted corner-case bugs, particularly when trying to use
      different SSL parameters for different connections.
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid corner-case memory leak in <application>libpq</> (Tom Lane)
     </para>

     <para>
      The reported problem involved leaking an error report
      during <function>PQreset()</>, but there might be related cases.
     </para>
    </listitem>

    <listitem>
     <para>
      Make <application>ecpg</>'s <option>--help</> and <option>--version</>
      options work consistently with our other executables (Haribabu Kommi)
     </para>
    </listitem>

    <listitem>
     <para>
      In <application>pg_dump</>, never dump range constructor functions
      (Tom Lane)
     </para>

     <para>
      This oversight led to <application>pg_upgrade</> failures with
      extensions containing range types, due to duplicate creation of the
      constructor functions.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <filename>contrib/intarray/bench/bench.pl</> to print the results
      of the <command>EXPLAIN</> it does when given the <option>-e</> option
      (Daniel Gustafsson)
     </para>
    </listitem>

    <listitem>
     <para>
      Update Windows time zone mapping to recognize some time zone names
      added in recent Windows versions (Michael Paquier)
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent failure of obsolete dynamic time zone abbreviations (Tom Lane)
     </para>

     <para>
      If a dynamic time zone abbreviation does not match any entry in the
      referenced time zone, treat it as equivalent to the time zone name.
      This avoids unexpected failures when IANA removes abbreviations from
      their time zone database, as they did in <application>tzdata</>
      release 2016f and seem likely to do again in the future.  The
      consequences were not limited to not recognizing the individual
      abbreviation; any mismatch caused
      the <structname>pg_timezone_abbrevs</> view to fail altogether.
     </para>
    </listitem>

    <listitem>
     <para>
      Update time zone data files to <application>tzdata</> release 2016h
      for DST law changes in Palestine and Turkey, plus historical
      corrections for Turkey and some regions of Russia.
      Switch to numeric abbreviations for some time zones in Antarctica,
      the former Soviet Union, and Sri Lanka.
     </para>

     <para>
      The IANA time zone database previously provided textual abbreviations
      for all time zones, sometimes making up abbreviations that have little
      or no currency among the local population.  They are in process of
      reversing that policy in favor of using numeric UTC offsets in zones
      where there is no evidence of real-world use of an English
      abbreviation.  At least for the time being, <productname>PostgreSQL</>
      will continue to accept such removed abbreviations for timestamp input.
      But they will not be shown in the <structname>pg_timezone_names</>
      view nor used for output.
     </para>

     <para>
      In this update, <literal>AMT</> is no longer shown as being in use to
      mean Armenia Time.  Therefore, we have changed the <literal>Default</>
      abbreviation set to interpret it as Amazon Time, thus UTC-4 not UTC+4.
     </para>
    </listitem>

   </itemizedlist>

  </sect2>
 </sect1>

 <sect1 id="release-9-2-18">
  <title>Release 9.2.18</title>

  <note>
  <title>Release Date</title>
  <simpara>2016-08-11</simpara>
  </note>

  <para>
   This release contains a variety of fixes from 9.2.17.
   For information about new features in the 9.2 major release, see
   <xref linkend="release-9-2">.
  </para>

  <sect2>
   <title>Migration to Version 9.2.18</title>

   <para>
    A dump/restore is not required for those running 9.2.X.
   </para>

   <para>
    However, if you are upgrading from a version earlier than 9.2.11,
    see <xref linkend="release-9-2-11">.
   </para>

  </sect2>

  <sect2>
   <title>Changes</title>

   <itemizedlist>

    <listitem>
     <para>
      Fix possible mis-evaluation of
      nested <literal>CASE</>-<literal>WHEN</> expressions (Heikki
      Linnakangas, Michael Paquier, Tom Lane)
     </para>

     <para>
      A <literal>CASE</> expression appearing within the test value
      subexpression of another <literal>CASE</> could become confused about
      whether its own test value was null or not.  Also, inlining of a SQL
      function implementing the equality operator used by
      a <literal>CASE</> expression could result in passing the wrong test
      value to functions called within a <literal>CASE</> expression in the
      SQL function's body.  If the test values were of different data
      types, a crash might result; moreover such situations could be abused
      to allow disclosure of portions of server memory.  (CVE-2016-5423)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix client programs' handling of special characters in database and
      role names (Noah Misch, Nathan Bossart, Michael Paquier)
     </para>

     <para>
      Numerous places in <application>vacuumdb</> and other client programs
      could become confused by database and role names containing double
      quotes or backslashes.  Tighten up quoting rules to make that safe.
      Also, ensure that when a conninfo string is used as a database name
      parameter to these programs, it is correctly treated as such throughout.
     </para>

     <para>
      Fix handling of paired double quotes
      in <application>psql</>'s <command>\connect</>
      and <command>\password</> commands to match the documentation.
     </para>

     <para>
      Introduce a new <option>-reuse-previous</> option
      in <application>psql</>'s <command>\connect</> command to allow
      explicit control of whether to re-use connection parameters from a
      previous connection.  (Without this, the choice is based on whether
      the database name looks like a conninfo string, as before.)  This
      allows secure handling of database names containing special
      characters in <application>pg_dumpall</> scripts.
     </para>

     <para>
      <application>pg_dumpall</> now refuses to deal with database and role
      names containing carriage returns or newlines, as it seems impractical
      to quote those characters safely on Windows.  In future we may reject
      such names on the server side, but that step has not been taken yet.
     </para>

     <para>
      These are considered security fixes because crafted object names
      containing special characters could have been used to execute
      commands with superuser privileges the next time a superuser
      executes <application>pg_dumpall</> or other routine maintenance
      operations.  (CVE-2016-5424)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix corner-case misbehaviors for <literal>IS NULL</>/<literal>IS NOT
      NULL</> applied to nested composite values (Andrew Gierth, Tom Lane)
     </para>

     <para>
      The SQL standard specifies that <literal>IS NULL</> should return
      TRUE for a row of all null values (thus <literal>ROW(NULL,NULL) IS
      NULL</> yields TRUE), but this is not meant to apply recursively
      (thus <literal>ROW(NULL, ROW(NULL,NULL)) IS NULL</> yields FALSE).
      The core executor got this right, but certain planner optimizations
      treated the test as recursive (thus producing TRUE in both cases),
      and <filename>contrib/postgres_fdw</> could produce remote queries
      that misbehaved similarly.
     </para>
    </listitem>

    <listitem>
     <para>
      Make the <type>inet</> and <type>cidr</> data types properly reject
      IPv6 addresses with too many colon-separated fields (Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent crash in <function>close_ps()</>
      (the <type>point</> <literal>##</> <type>lseg</> operator)
      for NaN input coordinates (Tom Lane)
     </para>

     <para>
      Make it return NULL instead of crashing.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix several one-byte buffer over-reads in <function>to_number()</>
      (Peter Eisentraut)
     </para>

     <para>
      In several cases the <function>to_number()</> function would read one
      more character than it should from the input string.  There is a
      small chance of a crash, if the input happens to be adjacent to the
      end of memory.
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid unsafe intermediate state during expensive paths
      through <function>heap_update()</> (Masahiko Sawada, Andres Freund)
     </para>

     <para>
      Previously, these cases locked the target tuple (by setting its XMAX)
      but did not WAL-log that action, thus risking data integrity problems
      if the page were spilled to disk and then a database crash occurred
      before the tuple update could be completed.
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid crash in <literal>postgres -C</> when the specified variable
      has a null string value (Michael Paquier)
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid consuming a transaction ID during <command>VACUUM</>
      (Alexander Korotkov)
     </para>

     <para>
      Some cases in <command>VACUUM</> unnecessarily caused an XID to be
      assigned to the current transaction.  Normally this is negligible,
      but if one is up against the XID wraparound limit, consuming more
      XIDs during anti-wraparound vacuums is a very bad thing.
     </para>
    </listitem>

    <listitem>
     <para>
      Avoid canceling hot-standby queries during <command>VACUUM FREEZE</>
      (Simon Riggs, &Aacute;lvaro Herrera)
     </para>

     <para>
      <command>VACUUM FREEZE</> on an otherwise-idle master server could
      result in unnecessary cancellations of queries on its standby
      servers.
     </para>
    </listitem>

    <listitem>
     <para>
      When a manual <command>ANALYZE</> specifies a column list, don't
      reset the table's <literal>changes_since_analyze</> counter
      (Tom Lane)
     </para>

     <para>
      If we're only analyzing some columns, we should not prevent routine
      auto-analyze from happening for the other columns.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <command>ANALYZE</>'s overestimation of <literal>n_distinct</>
      for a unique or nearly-unique column with many null entries (Tom
      Lane)
     </para>

     <para>
      The nulls could get counted as though they were themselves distinct
      values, leading to serious planner misestimates in some types of
      queries.
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent autovacuum from starting multiple workers for the same shared
      catalog (&Aacute;lvaro Herrera)
     </para>

     <para>
      Normally this isn't much of a problem because the vacuum doesn't take
      long anyway; but in the case of a severely bloated catalog, it could
      result in all but one worker uselessly waiting instead of doing
      useful work on other tables.
     </para>
    </listitem>

    <listitem>
     <para>
      Prevent infinite loop in GiST index build for geometric columns
      containing NaN component values (Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <filename>contrib/btree_gin</> to handle the smallest
      possible <type>bigint</> value correctly (Peter Eisentraut)
     </para>
    </listitem>

    <listitem>
     <para>
      Teach libpq to correctly decode server version from future servers
      (Peter Eisentraut)
     </para>

     <para>
      It's planned to switch to two-part instead of three-part server
      version numbers for releases after 9.6.  Make sure
      that <function>PQserverVersion()</> returns the correct value for
      such cases.
     </para>
    </listitem>

    <listitem>
     <para>
      Fix <application>ecpg</>'s code for <literal>unsigned long long</>
      array elements (Michael Meskes)
     </para>
    </listitem>

    <listitem>
     <para>
      In <application>pg_dump</> with both <option>-c</> and <option>-C</>
      options, avoid emitting an unwanted <literal>CREATE SCHEMA public</>
      command (David Johnston, Tom Lane)
     </para>
    </listitem>

    <listitem>
     <para>
      Make <application>pg_basebackup</> accept <literal>-Z 0</> as
      specifying no compression (Fujii Masao)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix makefiles' rule for building AIX shared libraries to be safe for
      parallel make (Noah Misch)
     </para>
    </listitem>

    <listitem>
     <para>
      Fix TAP tests and MSVC scripts to work when build directory's path
      name contains spaces (Michael Paquier, Kyotaro Horiguchi)
     </para>
    </listitem>

    <listitem>
     <para>
      Make regression tests safe for Danish and Welsh locales (Jeff Janes,
      Tom Lane)
     </para>

     <para>
      Change some test data that triggered the unusual sorting rules of
      these locales.
     </para>
    </listitem>

    <listitem>
     <para>
      Update our copy of the timezone code to match
      IANA's <application>tzcode</> release 2016c (Tom Lane)