From: Tom Lane <tgl@sss.pgh.pa.us>
To: 2365868844@qq.com
cc: pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #18770: The function 'packGraph ' causes a memory leak.
In-reply-to: <18770-895a62c1890d7491@postgresql.org>
References: <18770-895a62c1890d7491@postgresql.org>
Comments: In-reply-to PG Bug reporting form <noreply@postgresql.org>
	message dated "Fri, 10 Jan 2025 01:49:08 +0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <2421468.1736520496.1@sss.pgh.pa.us>
Date: Fri, 10 Jan 2025 09:48:16 -0500
Message-ID: <2421469.1736520496@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/2421469.1736520496%40sss.pgh.pa.us>
Precedence: bulk

PG Bug reporting form <noreply@postgresql.org> writes:
> In line 1965 of the trgm_regexp.c file
> 1965:	arcs = (TrgmPackArcInfo *)
> 1966:		palloc(sizeof(TrgmPackArcInfo) * trgmNFA->arcsCount);
> The memory requested 'arcs' was not freed.

This is not a bug, unless you can show that the memory is allocated
in a long-lived context that won't get freed reasonably soon.  We
generally assume that letting memory context cleanup recover memory
is both faster and more oversight-proof than retail pfree calls.

In this particular case, it looks to me like the memory will come from
the temp context made by createTrgmNFA, and therefore will be freed
before that returns.

			regards, tom lane