|
6 | 6 |
|
7 | 7 | <formalpara>
|
8 | 8 | <title>Release date:</title>
|
9 |
| - <para>2020-XX-XX, CURRENT AS OF 2020-05-03</para> |
| 9 | + <para>2020-XX-XX, CURRENT AS OF 2020-08-09</para> |
10 | 10 | </formalpara>
|
11 | 11 |
|
12 | 12 | <sect2>
|
|
21 | 21 | <itemizedlist>
|
22 | 22 |
|
23 | 23 | <listitem>
|
24 |
| - <para></para> |
| 24 | + <para>TBD</para> |
25 | 25 | </listitem>
|
26 | 26 |
|
27 | 27 | </itemizedlist>
|
|
141 | 141 |
|
142 | 142 | <listitem>
|
143 | 143 | <!--
|
144 |
| -Author: Peter Geoghegan < [email protected]> |
145 |
| -2020-03-07 [691e8b2e1] pageinspect: Fix types used for bt_metap() columns. |
146 |
| ---> |
147 |
| - |
148 |
| - <para> |
149 |
| - Fix <xref linkend="pageinspect"/>'s <function>bt_metap()</function> |
150 |
| - to return more appropriate data types that are less likely to overflow |
151 |
| - (Peter Geoghegan) |
152 |
| - </para> |
153 |
| - </listitem> |
154 |
| - |
155 |
| - <listitem> |
156 |
| -<!-- |
157 | 144 | Author: Fujii Masao < [email protected]>
|
158 | 145 | 2020-03-19 [1d253bae5] Rename the recovery-related wait events.
|
| 146 | +Author: Tom Lane < [email protected]> |
| 147 | +2020-05-15 [36ac359d3] Rename assorted LWLock tranches. |
| 148 | +2020-05-15 [474e7da64] Change locktype "speculative token" to "spectoken". |
| 149 | +2020-05-15 [14a910109] Drop the redundant "Lock" suffix from LWLock wait event |
| 150 | +2020-05-16 [3048898e7] Mop-up for wait event naming issues. |
159 | 151 | -->
|
160 | 152 |
|
161 | 153 | <para>
|
162 |
| - Rename some recovery-related <link linkend="wait-event-table">wait |
163 |
| - events</link> (Fujii Masao) |
164 |
| - </para> |
165 |
| - |
166 |
| - <para> |
167 |
| - Rename <literal>RecoveryWalAll</literal> |
168 |
| - to <literal>RecoveryWalStream</literal> |
169 |
| - and <literal>RecoveryWalStream</literal> to |
170 |
| - <literal>RecoveryRetrieveRetryInterval</literal>. |
| 154 | + Rename various <link linkend="wait-event-table">wait |
| 155 | + events</link> to improve consistency (Fujii Masao, Tom Lane) |
171 | 156 | </para>
|
172 | 157 | </listitem>
|
173 | 158 |
|
|
269 | 254 | </para>
|
270 | 255 | </listitem>
|
271 | 256 |
|
| 257 | + <listitem> |
| 258 | +<!-- |
| 259 | +Author: Tom Lane < [email protected]> |
| 260 | +2020-06-29 [21aac2ff9] Remove support for timezone "posixrules" file. |
| 261 | +--> |
| 262 | + |
| 263 | + <para> |
| 264 | + Remove support for <filename>posixrules</filename> files in the |
| 265 | + timezone database (Tom Lane) |
| 266 | + </para> |
| 267 | + |
| 268 | + <para> |
| 269 | + IANA's timezone group has deprecated this feature, meaning that it |
| 270 | + will gradually disappear from systems' timezone databases over the |
| 271 | + next few years. Rather than have a behavioral change appear |
| 272 | + unexpectedly with a timezone data update, we have |
| 273 | + removed <productname>PostgreSQL</productname>'s support for this |
| 274 | + feature as of version 13. This affects only the behavior |
| 275 | + of <link linkend="datetime-posix-timezone-specs">POSIX-style time |
| 276 | + zone specifications</link> that lack an explicit daylight savings |
| 277 | + transition rule; formerly the transition rule could be determined |
| 278 | + by installing a custom <filename>posixrules</filename> file, but |
| 279 | + now it is hard-wired. The recommended fix for any affected |
| 280 | + installations is to start using a geographical time zone name. |
| 281 | + </para> |
| 282 | + </listitem> |
| 283 | + |
| 284 | + <listitem> |
| 285 | +<!-- |
| 286 | +Author: Peter Geoghegan < [email protected]> |
| 287 | +2020-03-07 [691e8b2e1] pageinspect: Fix types used for bt_metap() columns. |
| 288 | +--> |
| 289 | + |
| 290 | + <para> |
| 291 | + Fix <xref linkend="pageinspect"/>'s <function>bt_metap()</function> |
| 292 | + to return more appropriate data types that are less likely to overflow |
| 293 | + (Peter Geoghegan) |
| 294 | + </para> |
| 295 | + </listitem> |
| 296 | + |
272 | 297 | </itemizedlist>
|
273 | 298 |
|
274 | 299 | </sect2>
|
|
607 | 632 | <!--
|
608 | 633 | Author: Tomas Vondra < [email protected]>
|
609 | 634 | 2020-04-06 [d2d8a229b] Implement Incremental Sort
|
| 635 | +Author: Peter Eisentraut < [email protected]> |
| 636 | +2020-07-05 [94e454cdd] Rename enable_incrementalsort for clarity |
610 | 637 | -->
|
611 | 638 |
|
612 | 639 | <para>
|
|
640 | 667 | 2020-03-18 [1f39bce02] Disk-based Hash Aggregation.
|
641 | 668 | Author: Jeff Davis < [email protected]>
|
642 | 669 | 2020-03-24 [dd8e19132] Consider disk-based hash aggregation to implement DISTIN
|
| 670 | +Author: Peter Geoghegan < [email protected]> |
| 671 | +2020-07-29 [78530c8e7] Add hash_mem_multiplier GUC. |
643 | 672 | -->
|
644 | 673 |
|
645 | 674 | <para>
|
646 | 675 | Allow <link linkend="guc-enable-hashagg">hash aggregation</link>
|
| 676 | + and <link linkend="queries-grouping-sets">grouping sets</link> |
647 | 677 | to use disk storage for large aggregation result sets (Jeff Davis)
|
648 | 678 | </para>
|
649 | 679 |
|
650 | 680 | <para>
|
651 | 681 | Previously, hash aggregation was avoided if it was expected to use
|
652 |
| - more than <xref linkend="guc-work-mem"/> memory. To reduce the |
653 |
| - likelihood of using disk storage for hash aggregation and attain |
654 |
| - behavior similar to previous Postgres releases, increase <xref |
655 |
| - linkend="guc-hash-mem-multiplier"/>. |
| 682 | + more than <xref linkend="guc-work-mem"/> memory. Now, a hash |
| 683 | + aggregation plan can be chosen despite that. The hash table will |
| 684 | + be spilled to disk if it exceeds <varname>work_mem</varname> times |
| 685 | + <xref linkend="guc-hash-mem-multiplier"/>. |
656 | 686 | </para>
|
657 |
| - </listitem> |
658 |
| - |
659 |
| - <listitem> |
660 |
| -<!-- |
661 |
| -Author: Jeff Davis < [email protected]> |
662 |
| -2020-03-18 [1f39bce02] Disk-based Hash Aggregation. |
663 |
| ---> |
664 | 687 |
|
665 | 688 | <para>
|
666 |
| - Allow <link linkend="queries-grouping-sets">grouping sets</link> to |
667 |
| - use hash aggregation with disk storage for large grouping set results |
668 |
| - (Jeff Davis) |
| 689 | + This behavior is normally preferable to the old behavior. But if |
| 690 | + it is inferior for a particular query, behavior similar to |
| 691 | + previous Postgres releases can be obtained by |
| 692 | + increasing <varname>hash_mem_multiplier</varname>. |
669 | 693 | </para>
|
670 | 694 | </listitem>
|
671 | 695 |
|
|
925 | 949 | <!--
|
926 | 950 | Author: Alvaro Herrera < [email protected]>
|
927 | 951 | 2019-11-08 [71a8a4f6e] Add backtrace support for error reporting
|
| 952 | +Author: Peter Eisentraut < [email protected]> |
| 953 | +2020-07-10 [8ff4d1277] Log the location field before any backtrace |
928 | 954 | -->
|
929 | 955 |
|
930 | 956 | <para>
|
|
1154 | 1180 | -->
|
1155 | 1181 |
|
1156 | 1182 | <para>
|
1157 |
| - Change the default minimum <acronym>TLS</acronym> version from 1.0 |
1158 |
| - to 1.2 (Peter Eisentraut) |
| 1183 | + Change the server's default minimum <acronym>TLS</acronym> version |
| 1184 | + for encrypted connections from 1.0 to 1.2 (Peter Eisentraut) |
1159 | 1185 | </para>
|
1160 | 1186 |
|
1161 | 1187 | <para>
|
|
1984 | 2010 | -->
|
1985 | 2011 |
|
1986 | 2012 | <para>
|
1987 |
| - Allow libpq clients to require channel binding (Jeff Davis) |
| 2013 | + Allow libpq clients to require channel binding for encrypted |
| 2014 | + connections (Jeff Davis) |
1988 | 2015 | </para>
|
1989 | 2016 |
|
1990 | 2017 | <para>
|
|
2001 | 2028 | 2020-01-28 [ff8ca5fad] Add connection parameters to control SSL protocol min/ma
|
2002 | 2029 | Author: Michael Paquier < [email protected]>
|
2003 | 2030 | 2020-04-30 [401aad670] Rename connection parameters to control min/max SSL prot
|
| 2031 | +Author: Tom Lane < [email protected]> |
| 2032 | +2020-06-27 [16412c784] Change libpq's default ssl_min_protocol_version to TLSv1 |
2004 | 2033 | -->
|
2005 | 2034 |
|
2006 | 2035 | <para>
|
2007 | 2036 | Add libpq connection parameters to control the min/max
|
2008 |
| - <acronym>TLS</acronym> version (Daniel Gustafsson) |
| 2037 | + <acronym>TLS</acronym> version for encrypted connections |
| 2038 | + (Daniel Gustafsson) |
2009 | 2039 | </para>
|
2010 | 2040 |
|
2011 | 2041 | <para>
|
2012 | 2042 | The settings are <xref
|
2013 | 2043 | linkend="libpq-connect-ssl-min-protocol-version"/> and <xref
|
2014 | 2044 | linkend="libpq-connect-ssl-max-protocol-version"/>.
|
| 2045 | + By default, the minimum <acronym>TLS</acronym> version is 1.2 |
| 2046 | + (this represents a behavioral change from previous releases). |
2015 | 2047 | </para>
|
2016 | 2048 | </listitem>
|
2017 | 2049 |
|
|
2022 | 2054 | -->
|
2023 | 2055 |
|
2024 | 2056 | <para>
|
2025 |
| - Tighten line length and comment detection in <link |
| 2057 | + Tighten libpq's overlength-line handling and comment detection for <link |
2026 | 2058 | linkend="libpq-pgpass">.pgpass</link> files (Fujii Masao)
|
2027 | 2059 | </para>
|
2028 | 2060 | </listitem>
|
|
2056 | 2088 | </para>
|
2057 | 2089 | </listitem>
|
2058 | 2090 |
|
| 2091 | + <listitem> |
| 2092 | +<!-- |
| 2093 | +Author: Tom Lane < [email protected]> |
| 2094 | +2020-08-03 [44cd434ec] Fix behavior of ecpg's "EXEC SQL elif name". |
| 2095 | +--> |
| 2096 | + |
| 2097 | + <para> |
| 2098 | + Fix <application>ecpg</application>'s <literal>EXEC SQL |
| 2099 | + elif</literal> directive to work correctly (Tom Lane) |
| 2100 | + </para> |
| 2101 | + |
| 2102 | + <para> |
| 2103 | + Previously it behaved the same as <literal>endif</literal> followed |
| 2104 | + by <literal>ifdef</literal>, so that a successful previous branch |
| 2105 | + of the same <literal>if</literal> construct did not prevent |
| 2106 | + expansion of the <literal>elif</literal> branch or following |
| 2107 | + branches. |
| 2108 | + </para> |
| 2109 | + </listitem> |
| 2110 | + |
2059 | 2111 | </itemizedlist>
|
2060 | 2112 |
|
2061 | 2113 | </sect3>
|
|
2798 | 2850 | </para>
|
2799 | 2851 | </listitem>
|
2800 | 2852 |
|
| 2853 | + <listitem> |
| 2854 | +<!-- |
| 2855 | +Author: Tom Lane < [email protected]> |
| 2856 | +2020-07-24 [92fe6895d] Fix assorted bugs by changing TS_execute's callback API |
| 2857 | +2020-07-24 [70eca6a9a] Replace TS_execute's TS_EXEC_CALC_NOT flag with TS_EXEC_ |
| 2858 | +--> |
| 2859 | + |
| 2860 | + <para> |
| 2861 | + Change the API for <function>TS_execute()</function> (Tom Lane, |
| 2862 | + Pavel Borisov) |
| 2863 | + </para> |
| 2864 | + |
| 2865 | + <para> |
| 2866 | + <function>TS_execute</function> callbacks must now provide ternary |
| 2867 | + (yes/no/maybe) logic. Calculating NOT queries accurately is now |
| 2868 | + the default. |
| 2869 | + </para> |
| 2870 | + </listitem> |
| 2871 | + |
2801 | 2872 | </itemizedlist>
|
2802 | 2873 |
|
2803 | 2874 | </sect3>
|
|
2915 | 2986 | <!--
|
2916 | 2987 | Author: Fujii Masao < [email protected]>
|
2917 | 2988 | 2020-04-02 [17e032822] Allow pg_stat_statements to track planning statistics.
|
| 2989 | +2020-07-03 [8d459762b] Change default of pg_stat_statements.track_planning to o |
2918 | 2990 | -->
|
2919 | 2991 |
|
2920 | 2992 | <para>
|
|
0 commit comments