Fix failure of contrib/auto_explain to print per-node timing information.

This has been broken since commit af7914c662,
which added the EXPLAIN (TIMING) option.  Although that commit included
updates to auto_explain, they evidently weren't tested very carefully,
because the code failed to print node timings even when it should, due to
failure to set es.timing in the ExplainState struct.  Reported off-list by
Neelakanth Nadgir of Salesforce.

In passing, clean up the documentation for auto_explain's options a
little bit, including re-ordering them into what seems to me a more
logical order.
This commit is contained in:
Tom Lane 2014-09-19 13:18:56 -04:00
parent a92b5f969b
commit 898f8a96ef
2 changed files with 61 additions and 51 deletions

View File

@ -26,7 +26,7 @@ static bool auto_explain_log_analyze = false;
static bool auto_explain_log_verbose = false;
static bool auto_explain_log_buffers = false;
static bool auto_explain_log_triggers = false;
static bool auto_explain_log_timing = false;
static bool auto_explain_log_timing = true;
static int auto_explain_log_format = EXPLAIN_FORMAT_TEXT;
static bool auto_explain_log_nested_statements = false;
@ -200,8 +200,6 @@ explain_ExecutorStart(QueryDesc *queryDesc, int eflags)
queryDesc->instrument_options |= INSTRUMENT_TIMER;
else
queryDesc->instrument_options |= INSTRUMENT_ROWS;
if (auto_explain_log_buffers)
queryDesc->instrument_options |= INSTRUMENT_BUFFERS;
}
@ -302,6 +300,7 @@ explain_ExecutorEnd(QueryDesc *queryDesc)
es.analyze = (queryDesc->instrument_options && auto_explain_log_analyze);
es.verbose = auto_explain_log_verbose;
es.buffers = (es.analyze && auto_explain_log_buffers);
es.timing = (es.analyze && auto_explain_log_timing);
es.format = auto_explain_log_format;
ExplainBeginOutput(&es);

View File

@ -81,28 +81,13 @@ LOAD 'auto_explain';
When this parameter is on, per-plan-node timing occurs for all
statements executed, whether or not they run long enough to actually
get logged. This can have an extremely negative impact on performance.
Turning off <varname>auto_explain.log_timing</varname> ameliorates the
performance cost, at the price of obtaining less information.
</para>
</note>
</listitem>
</varlistentry>
<varlistentry>
<term>
<varname>auto_explain.log_verbose</varname> (<type>boolean</type>)
<indexterm>
<primary><varname>auto_explain.log_verbose</> configuration parameter</primary>
</indexterm>
</term>
<listitem>
<para>
<varname>auto_explain.log_verbose</varname> causes <command>EXPLAIN VERBOSE</>
output, rather than just <command>EXPLAIN</> output, to be printed
when an execution plan is logged. This parameter is off by default.
Only superusers can change this setting.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<varname>auto_explain.log_buffers</varname> (<type>boolean</type>)
@ -112,12 +97,37 @@ LOAD 'auto_explain';
</term>
<listitem>
<para>
<varname>auto_explain.log_buffers</varname> causes <command>EXPLAIN
(ANALYZE, BUFFERS)</> output, rather than just <command>EXPLAIN</>
output, to be printed when an execution plan is logged. This parameter is
off by default. Only superusers can change this setting. This
parameter has no effect unless <varname>auto_explain.log_analyze</>
parameter is set.
<varname>auto_explain.log_buffers</varname> controls whether buffer
usage statistics are printed when an execution plan is logged; it's
equivalent to the <literal>BUFFERS</> option of <command>EXPLAIN</>.
This parameter has no effect
unless <varname>auto_explain.log_analyze</varname> is enabled.
This parameter is off by default.
Only superusers can change this setting.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<varname>auto_explain.log_timing</varname> (<type>boolean</type>)
<indexterm>
<primary><varname>auto_explain.log_timing</> configuration parameter</primary>
</indexterm>
</term>
<listitem>
<para>
<varname>auto_explain.log_timing</varname> controls whether per-node
timing information is printed when an execution plan is logged; it's
equivalent to the <literal>TIMING</> option of <command>EXPLAIN</>.
The overhead of repeatedly reading the system clock can slow down
queries significantly on some systems, so it may be useful to set this
parameter to off when only actual row counts, and not exact times, are
needed.
This parameter has no effect
unless <varname>auto_explain.log_analyze</varname> is enabled.
This parameter is on by default.
Only superusers can change this setting.
</para>
</listitem>
</varlistentry>
@ -133,9 +143,28 @@ LOAD 'auto_explain';
<para>
<varname>auto_explain.log_triggers</varname> causes trigger
execution statistics to be included when an execution plan is logged.
This parameter is off by default. Only superusers can change this
setting. This parameter has no effect unless
<varname>auto_explain.log_analyze</> parameter is set.
This parameter has no effect
unless <varname>auto_explain.log_analyze</varname> is enabled.
This parameter is off by default.
Only superusers can change this setting.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<varname>auto_explain.log_verbose</varname> (<type>boolean</type>)
<indexterm>
<primary><varname>auto_explain.log_verbose</> configuration parameter</primary>
</indexterm>
</term>
<listitem>
<para>
<varname>auto_explain.log_verbose</varname> controls whether verbose
details are printed when an execution plan is logged; it's
equivalent to the <literal>VERBOSE</> option of <command>EXPLAIN</>.
This parameter is off by default.
Only superusers can change this setting.
</para>
</listitem>
</varlistentry>
@ -158,27 +187,6 @@ LOAD 'auto_explain';
</listitem>
</varlistentry>
<varlistentry>
<term>
<varname>auto_explain.log_timing</varname> (<type>boolean</type>)
<indexterm>
<primary><varname>auto_explain.log_timing</> configuration parameter</primary>
</indexterm>
</term>
<listitem>
<para>
<varname>auto_explain.log_timing</varname> causes <command>EXPLAIN
(ANALYZE, TIMING off)</> output, rather than just <command>EXPLAIN (ANALYZE)</>
output. The overhead of repeatedly reading the system clock can slow down the
query significantly on some systems, so it may be useful to set this
parameter to off when only actual row counts, and not exact times, are needed.
This parameter is only effective when <varname>auto_explain.log_analyze</varname>
is also enabled. This parameter is on by default.
Only superusers can change this setting.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<varname>auto_explain.log_nested_statements</varname> (<type>boolean</type>)
@ -198,7 +206,9 @@ LOAD 'auto_explain';
</variablelist>
<para>
These parameters must be set in <filename>postgresql.conf</>.
In ordinary usage, these parameters are set
in <filename>postgresql.conf</>, although superusers can alter them
on-the-fly within their own sessions.
Typical usage might be:
</para>
@ -216,6 +226,7 @@ auto_explain.log_min_duration = '3s'
<programlisting>
postgres=# LOAD 'auto_explain';
postgres=# SET auto_explain.log_min_duration = 0;
postgres=# SET auto_explain.log_analyze = true;
postgres=# SELECT count(*)
FROM pg_class, pg_index
WHERE oid = indrelid AND indisunique;