RE: [HACKERS] freefuncs.c is never called from anywhere!?

Поиск
Список
Период
Сортировка
От Hiroshi Inoue
Тема RE: [HACKERS] freefuncs.c is never called from anywhere!?
Дата
Msg-id 001101bf6b7e$c948b560$2801007e@tpf.co.jp
обсуждение исходный текст
Ответ на Re: [HACKERS] freefuncs.c is never called from anywhere!?  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: [HACKERS] freefuncs.c is never called from anywhere!?  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
> -----Original Message-----
> From: Tom Lane [mailto:tgl@sss.pgh.pa.us]
> 
> "Hiroshi Inoue" <Inoue@tpf.co.jp> writes:
> > As far as I see,freeObject() has a fundamental problem.
> > Probably it couldn't free multiple references safely.
> 
> Yes, that code will have a problem with node trees that have
> circular references or multiple links to the same node.
> Both of which are cases that we can currently generate :-(.
> 
> Of course, circular links will also break copyfuncs, equalfuncs,
> and printfuncs.  We have a known bug with the backend crashing
> when -d is enabled for certain queries, because of infinite
> recursion in printfuncs.
>

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() ?

Regards.

Hiroshi Inoue
Inoue@tpf.co.jp


В списке pgsql-hackers по дате отправления:

Предыдущее
От: "Hiroshi Inoue"
Дата:
Сообщение: RE: ImmediateSharedRelationCacheInvalidate considered harmful
Следующее
От: Chris Bitmead
Дата:
Сообщение: Re: [HACKERS] Re: ORDBMS