From: Tom Lane <tgl@sss.pgh.pa.us>
To: Pailloncy Jean-Gerard <jg@rilk.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Memory leak tsearch2 VACUUM FULL VERBOSE ANALYZE
Date: Thu, 23 Dec 2004 17:44:34 -0500
Message-ID: <2035.1103841874@sss.pgh.pa.us> (raw)
In-Reply-To: <233CAA75-5473-11D9-9A44-000A95DE2550@rilk.com>
References: <3D4B6E62-5055-11D9-96E2-000A95DE2550@ifrance.com>
<DFED184A-5076-11D9-96E2-000A95DE2550@rilk.com>
<24692.1103322047@sss.pgh.pa.us>
<233CAA75-5473-11D9-9A44-000A95DE2550@rilk.com>
Pailloncy Jean-Gerard <jg@rilk.com> writes:
> I think I have a test case for 7.4.2
Try the attached patch.
It looked to me like there were some smaller leaks going on during COPY
and CREATE INDEX, which I will look into later --- but this seems to be
the problem for VACUUM FULL.
regards, tom lane
inline patch 1: vacuum.c (+2/-0 lines) download inline-1.patchIndex: vacuum.c
===================================================================
RCS file: /cvsroot/pgsql/src/backend/commands/vacuum.c,v
retrieving revision 1.263
diff -c -r1.263 vacuum.c
*** vacuum.c 2 Oct 2003 23:19:44 -0000 1.263
--- vacuum.c 23 Dec 2004 22:37:57 -0000
***************
*** 2041,2046 ****
--- 2041,2047 ----
ExecStoreTuple(&newtup, slot, InvalidBuffer, false);
ExecInsertIndexTuples(slot, &(newtup.t_self),
estate, true);
+ ResetPerTupleExprContext(estate);
}
WriteBuffer(cur_buffer);
***************
*** 2174,2179 ****
--- 2175,2181 ----
{
ExecStoreTuple(&newtup, slot, InvalidBuffer, false);
ExecInsertIndexTuples(slot, &(newtup.t_self), estate, true);
+ ResetPerTupleExprContext(estate);
}
} /* walk along page */
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: tgl@sss.pgh.pa.us, jg@rilk.com
Subject: Re: Memory leak tsearch2 VACUUM FULL VERBOSE ANALYZE
In-Reply-To: <2035.1103841874@sss.pgh.pa.us>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox