database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-12-24 06:01:07 +03:00
Code Activity

Move isolationtester's is-blocked query into C code for speed.

Browse Source 
Commit 4deb41381 modified isolationtester's query to see whether a
session is blocked to also check for waits occurring in GetSafeSnapshot.
However, it did that in a way that enormously increased the query's
runtime under CLOBBER_CACHE_ALWAYS, causing the buildfarm members
that use that to run about four times slower than before, and in some
cases fail entirely.  To fix, push the entire logic into a dedicated
backend function.  This should actually reduce the CLOBBER_CACHE_ALWAYS
runtime from what it was previously, though I've not checked that.

In passing, expose a SQL function to check for safe-snapshot blockage,
comparable to pg_blocking_pids.  This is more or less free given the
infrastructure built to solve the other problem, so we might as well.

Thomas Munro

Discussion: https://postgr.es/m/20170407165749.pstcakbc637opkax@alap3.anarazel.de
This commit is contained in:
Tom Lane
2017-04-10 10:26:54 -04:00
parent 9cf5c31964
commit 511540dadf
8 changed files with 206 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
doc/src/sgml/func.sgml
View File

@@ -15747,7 +15747,7 @@ SELECT * FROM pg_ls_dir('.') WITH ORDINALITY AS t(ls,n);
<row>
<entry><literal><function>pg_blocking_pids(<type>int</type>)</function></literal></entry>
<entry><type>int[]</type></entry>
<entry>Process ID(s) that are blocking specified server process ID</entry>
<entry>Process ID(s) that are blocking specified server process ID from acquiring a lock</entry>
</row>
<row>
@@ -15793,6 +15793,12 @@ SELECT * FROM pg_ls_dir('.') WITH ORDINALITY AS t(ls,n);
<entry>server start time</entry>
</row>
<row>
<entry><literal><function>pg_safe_snapshot_blocking_pids(<type>int</type>)</function></literal></entry>
<entry><type>int[]</type></entry>
<entry>Process ID(s) that are blocking specified server process ID from acquiring a safe snapshot</entry>
</row>
<row>
<entry><literal><function>pg_trigger_depth()</function></literal></entry>
<entry><type>int</type></entry>
@@ -16067,6 +16073,25 @@ SET search_path TO <replaceable>schema</> <optional>, <replaceable>schema</>, ..
server started.
</para>
<indexterm>
<primary>pg_safe_snapshot_blocking_pids</primary>
</indexterm>
<para>
<function>pg_safe_snapshot_blocking_pids</function> returns an array of
the process IDs of the sessions that are blocking the server process with
the specified process ID from acquiring a safe snapshot, or an empty array
if there is no such server process or it is not blocked. A session
running a <literal>SERIALIZABLE</literal> transaction blocks
a <literal>SERIALIZABLE READ ONLY DEFERRABLE</literal> transaction from
acquiring a snapshot until the latter determines that it is safe to avoid
taking any predicate locks. See <xref linkend="xact-serializable"> for
more information about serializable and deferrable transactions. Frequent
calls to this function could have some impact on database performance,
because it needs access to the predicate lock manager's shared
state for a short time.
</para>
<indexterm>
<primary>version</primary>
</indexterm>