Skip to content

P
Postgres FD Implementation
Commit 61b96234 authored by Bruce Momjian's avatar Bruce Momjian
Browse files
Options

Reorder some 9.3 release item entries 

More to go.
parent 22f956f5
Show whitespace changes
Inline Side-by-side
Showing with 149 additions and 161 deletions
doc/src/sgml/release-9.3.sgml
View file @ 61b96234
...@@ -60,79 +60,59 @@ ...@@ -60,79 +60,59 @@
<title>Server</title> <title>Server</title>
<sect4> <sect4>
<title>Performance</title> <title>Locking</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
Improve performance of NUMERIC calculations (Kyotaro Horiguchi) Prevent non-key-field row updates from locking foreign key rows
</para> (AlvaroAacute;lvaro Herrera, Marti Raudsepp, Alexander Shulgin,
</listitem> Noah Misch and Andres Freund)
<listitem>
<para>
Use a binary heap for Merge-Append operations (Abhijit Menon-Sen)
</para> </para>
</listitem>
<listitem>
<para> <para>
Add COPY FREEZE option to avoid the overhead of later marking This improves concurrency and reduces the probability of deadlocks.
tuples as committed (Simon Riggs, Jeff Davis) UPDATEs on non-key columns use the new SELECT FOR NO KEY UPDATE lock
type, and foreign key checks use the new SELECT FOR KEY SHARE lock mode.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Improve memory usage for in-memory sorts (Jeff Janes) Add cache of local locks (Jeff Janes)
</para> </para>
<para> <para>
Users who have set work_mem based on the previous behavior should This speeds lock release at statement completion in transactions
revisit that setting. that hold many locks; it is particularly useful for pg_dump.
</para> </para>
</listitem> </listitem>
<listitem> </itemizedlist>
<para>
Improve performance for transactions creating, rebuilding, or
dropping many relations (Jeff Janes, Tomas Vondra)
</para>
</listitem>
<listitem> </sect4>
<para>
Improve performance of the CREATE TABLE ... ON COMMIT DELETE ROWS <sect4>
clause by only issuing delete if the temporary table was accessed <title>Indexes</title>
(Heikki Linnakangas)
</para> <itemizedlist>
</listitem>
<listitem> <listitem>
<para> <para>
Allow heap-only tuple updates on system tables (Andres Freund) Add SP-GiST support for range data types (Alexander Korotkov)
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Have vacuum recheck visibility after it has removed expired tuples Allow unlogged GiST indexes (Jeevan Chalke)
(Pavan Deolasee)
</para>
<para>
This increases the chance of a page being marked as all-visible.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Split pgstat file in per-database and global files (Tomas Vondra) Improve concurrency of hash indexes (Robert Haas)
</para>
<para>
This reduces the statistics management read and write overhead.
</para> </para>
</listitem> </listitem>
...@@ -147,8 +127,8 @@ ...@@ -147,8 +127,8 @@
<listitem> <listitem>
<para> <para>
Reduce optimizer overhead by discarding plans with unneeded cheaper Improve the ability to detect indexable prefixes in regular
startup costs (Tom Lane) expressions (Tom Lane)
</para> </para>
</listitem> </listitem>
...@@ -161,8 +141,8 @@ ...@@ -161,8 +141,8 @@
<listitem> <listitem>
<para> <para>
Improve the ability to detect indexable prefixes in regular Reduce optimizer overhead by discarding plans with unneeded cheaper
expressions (Tom Lane) startup costs (Tom Lane)
</para> </para>
</listitem> </listitem>
...@@ -177,72 +157,79 @@ ...@@ -177,72 +157,79 @@
</sect4> </sect4>
<sect4> <sect4>
<title>Indexes</title> <title>General Performance</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
Reduce use of heavyweight locking inside hash AM. Add COPY FREEZE option to avoid the overhead of later marking
tuples as committed (Simon Riggs, Jeff Davis)
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Improve concurrency of hash indexes (Robert Haas) Improve performance of NUMERIC calculations (Kyotaro Horiguchi)
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Add SP-GiST support for range data types (Alexander Korotkov) Use a binary heap for Merge-Append operations (Abhijit Menon-Sen)
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Allow unlogged GiST indexes (Jeevan Chalke) Improve memory usage for in-memory sorts (Jeff Janes)
</para>
<para>
Users who have set work_mem based on the previous behavior should
revisit that setting.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Allow SP-GiST to support of the range adjacent operator (Alexander Improve performance for transactions creating, rebuilding, or
Korotkov) dropping many relations (Jeff Janes, Tomas Vondra)
</para> </para>
</listitem> </listitem>
</itemizedlist> <listitem>
<para>
</sect4> Improve performance of the CREATE TABLE ... ON COMMIT DELETE ROWS
clause by only issuing delete if the temporary table was accessed
<sect4> (Heikki Linnakangas)
<title>Locking</title> </para>
</listitem>
<itemizedlist> <listitem>
<para>
Allow heap-only tuple updates on system tables (Andres Freund)
</para>
</listitem>
<listitem> <listitem>
<para> <para>
Prevent non-key-field row updates from locking foreign key rows Have vacuum recheck visibility after it has removed expired tuples
(AlvaroAacute;lvaro Herrera, Marti Raudsepp, Alexander Shulgin, (Pavan Deolasee)
Noah Misch and Andres Freund)
</para> </para>
<para> <para>
This improves concurrency and reduces the probability of deadlocks. This increases the chance of a page being marked as all-visible.
UPDATEs on non-key columns use the new SELECT FOR NO KEY UPDATE lock
type, and foreign key checks use the new SELECT FOR KEY SHARE lock mode.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Add cache of local locks (Jeff Janes) Split pgstat file in per-database and global files (Tomas Vondra)
</para> </para>
<para> <para>
This speeds lock release at statement completion in transactions This reduces the statistics management read and write overhead.
that hold many locks; it is particularly useful for pg_dump.
</para> </para>
</listitem> </listitem>
...@@ -251,42 +238,52 @@ ...@@ -251,42 +238,52 @@
</sect4> </sect4>
<sect4> <sect4>
<title>Authentication</title> <title>Monitoring</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
Improve LDAP error reporting and documentation (Peter Eisentraut) Add optional ability to checksum data pages and report corruption
(Simon Riggs, Jeff Davis, Greg Smith)
</para>
<para>
The checksum option can be set during initdb.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Add support for LDAP authentication to be specified in URL format Allow pg_terminate_backend() to terminate other backends with
(Peter Eisentraut) the same role (Dan Farina)
</para>
<para>
Previously, only superusers could terminate other sessions.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Change the ssl_ciphers parameter to start with DEFAULT, rather Allow the statistics collector to operate properly in cases where
than ALL, then remove insecure ciphers (Magnus Hagander) the system clock goes backwards (Tom Lane)
</para> </para>
<para> <para>
It is assumed DEFAULT is more appropriate cipher set. Previously statistics collection would stop until the time again
reached the previously-stored latest time.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Parse/load pg_ident.conf once, not during each connection (Amit Add function to report the size of the GIN pending index insertion
Kapila) list (Fujii Masao)
</para> </para>
<para> <para>
This is similar to how pg_hba.conf is processed. The function is pgstatginindex() in pgstattuple.
</para> </para>
</listitem> </listitem>
...@@ -295,52 +292,42 @@ ...@@ -295,52 +292,42 @@
</sect4> </sect4>
<sect4> <sect4>
<title>Monitoring</title> <title>Authentication</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
Allow the statistics collector to operate properly in cases where Improve LDAP error reporting and documentation (Peter Eisentraut)
the system clock goes backwards (Tom Lane)
</para>
<para>
Previously statistics collection would stop until the time again
reached the previously-stored latest time.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Add optional ability to checksum data pages and report corruption Add support for LDAP authentication to be specified in URL format
(Simon Riggs, Jeff Davis, Greg Smith) (Peter Eisentraut)
</para>
<para>
The checksum option can be set during initdb.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Add function to report the size of the GIN pending index insertion Change the ssl_ciphers parameter to start with DEFAULT, rather
list (Fujii Masao) than ALL, then remove insecure ciphers (Magnus Hagander)
</para> </para>
<para> <para>
The function is pgstatginindex() in pgstattuple. It is assumed DEFAULT is more appropriate cipher set.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Allow pg_terminate_backend() to terminate other backends with Parse/load pg_ident.conf once, not during each connection (Amit
the same role (Dan Farina) Kapila)
</para> </para>
<para> <para>
Previously, only superusers could terminate other sessions. This is similar to how pg_hba.conf is processed.
</para> </para>
</listitem> </listitem>
...@@ -367,44 +354,43 @@ ...@@ -367,44 +354,43 @@
<listitem> <listitem>
<para> <para>
Increase the maximum initdb-configured value for shared_buffers Allow the postmaster to listen on multiple Unix-domain sockets
to 128MB (Robert Haas) (Honza Horak)
</para> </para>
<para> <para>
This is the maximum value initdb attempts to set in This renames configuration parameter unix_socket_directory to
postgresql.conf; the previous value was 32MB. unix_socket_directories, which accepts a list of directories.
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Allow a directory of configuration files to be processed (Magnus Add configuration variable lock_timeout to limit lock wait duration
Hagander, Greg Smith, Selena Deckelmann) (Zoltán Böszörményi)
</para> </para>
</listitem>
<listitem>
<para> <para>
The directory is specified as include_dir in server configuration file. Allow a directory of configuration files to be processed (Magnus
Hagander, Greg Smith, Selena Deckelmann)
</para> </para>
</listitem>
<listitem>
<para> <para>
Have "session id" in log_line_prefix (%c) always output four hex The directory is specified as include_dir in server configuration file.
digits after the period (Bruce Momjian)
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Add configuration variable lock_timeout to limit lock wait duration Increase the maximum initdb-configured value for shared_buffers
(Zoltán Böszörményi) to 128MB (Robert Haas)
</para> </para>
</listitem>
<listitem>
<para> <para>
Remove the external PID file on postmaster exit (Peter Eisentraut) This is the maximum value initdb attempts to set in
postgresql.conf; the previous value was 32MB.
</para> </para>
</listitem> </listitem>
...@@ -427,15 +413,17 @@ ...@@ -427,15 +413,17 @@
</para> </para>
</listitem> </listitem>
<listitem> <listitem>
<para> <para>
Allow the postmaster to listen on multiple Unix-domain sockets Remove the external PID file on postmaster exit (Peter Eisentraut)
(Honza Horak)
</para> </para>
</listitem>
<listitem>
<para> <para>
This renames configuration parameter unix_socket_directory to Have "session id" in log_line_prefix (%c) always output four hex
unix_socket_directories, which accepts a list of directories. digits after the period (Bruce Momjian) COMPATIBILITY
</para> </para>
</listitem> </listitem>
...@@ -450,6 +438,18 @@ ...@@ -450,6 +438,18 @@
<itemizedlist> <itemizedlist>
<listitem>
<para>
Allow a streaming replication standbys to follow a timeline switch (Heikki Linnakangas)
</para>
<para>
This allows streaming standbys to feed from newly-promoted slaves.
Previously slaves required access to a WAL archive directory to
accomplish this.
</para>
</listitem>
<listitem> <listitem>
<para> <para>
Add SQL functions pg_backup_in_progress() and pg_backup_start_time() (Darold Gilles) Add SQL functions pg_backup_in_progress() and pg_backup_start_time() (Darold Gilles)
...@@ -485,18 +485,6 @@ ...@@ -485,18 +485,6 @@
</para> </para>
</listitem> </listitem>
<listitem>
<para>
Allow a streaming replication standbys to follow a timeline switch (Heikki Linnakangas)
</para>
<para>
This allows streaming standbys to feed from newly-promoted slaves.
Previously slaves required access to a WAL archive directory to
accomplish this.
</para>
</listitem>
<listitem> <listitem>
<para> <para>
Have pg_basebackup --write-recovery-conf output a minimal Have pg_basebackup --write-recovery-conf output a minimal
...@@ -678,6 +666,37 @@ ...@@ -678,6 +666,37 @@
</itemizedlist> </itemizedlist>
<sect4>
<title><link linkend="SQL-CREATETABLE"><command>CREATE TABLE</></link></title>
<itemizedlist>
<listitem>
<para>
No longer output messages about implicit index and sequence creation (Robert Haas)
</para>
<para>
These messages now appear with DEBUG1-level output.
</para>
</listitem>
<listitem>
<para>
Allow CREATE TABLE to succeed for a non-existent schema
(Bruce Momjian)
</para>
<para>
Previously, CREATE TABLE IF EXIST threw an error if the schema was
nonexistent.
</para>
</listitem>
</itemizedlist>
</sect4>
<sect4> <sect4>
<title>Constraints</title> <title>Constraints</title>
...@@ -822,37 +841,6 @@ ...@@ -822,37 +841,6 @@
</sect4> </sect4>
<sect4>
<title><link linkend="SQL-CREATETABLE"><command>CREATE TABLE</></link></title>
<itemizedlist>
<listitem>
<para>
No longer output messages about implicit index and sequence creation (Robert Haas)
</para>
<para>
These messages now appear with DEBUG1-level output.
</para>
</listitem>
<listitem>
<para>
Allow CREATE TABLE to succeed for a non-existent schema
(Bruce Momjian)
</para>
<para>
Previously, CREATE TABLE IF EXIST threw an error if the schema was
nonexistent.
</para>
</listitem>
</itemizedlist>
</sect4>
</sect3> </sect3>
<sect3> <sect3>
...@@ -1698,7 +1686,7 @@ ...@@ -1698,7 +1686,7 @@
<listitem> <listitem>
<para> <para>
Allow pgbench to use scale values larger than 21474 (Greg Smith) Allow pgbench to use a larger scale factor (Greg Smith)
</para> </para>
</listitem> </listitem>
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment