Re: [HACKERS] freefuncs.c is never called from anywhere!? - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [HACKERS] freefuncs.c is never called from anywhere!?
Date
Msg-id 12665.949283714@sss.pgh.pa.us
Whole thread Raw
In response to RE: [HACKERS] freefuncs.c is never called from anywhere!?  ("Hiroshi Inoue" <Inoue@tpf.co.jp>)
List pgsql-hackers
"Hiroshi Inoue" <Inoue@tpf.co.jp> writes:
> Multiple links to the same node is not so fatal for other xxxxObject().
> In fact they are used without big problem.
> But isn't it fatal for freeObject() ?

Yes, of course.  We could possibly fix that by maintaining a list of
already-freed nodes during any one freeObject() call, but that seems
painful and slow.

However, the case I'm interested in would only be trying to free
nodetrees that were created by either copyObject or stringToNode,
and since neither of those routines produce multiple or circular
links, it seems safe enough to use freeObject as-is for the purpose.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: [HACKERS] Another nasty cache problem
Next
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Another nasty cache problem