From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
Subject: | Re: Fixing code that ignores failure of XLogRecGetBlockTag |
Date: | 2022-04-11 20:57:47 |
Message-ID: | CA+TgmoaOdNdd3fBsM0U=OqDrxJZST3z0ayA4dqVVKhXVoZ0vQQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mon, Apr 11, 2022 at 2:20 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Currently, XLogRecGetBlockTag has 41 callers, of which only four
> bother to check the function's result. The remainder take it on
> faith that they got valid data back, and many of them will
> misbehave in seriously nasty ways if they didn't. (This point
> was drawn to my attention by a Coverity complaint.)
>
> I think we should make this a little less fragile. Since we
> already have XLogRecGetBlockTagExtended, I propose that callers
> that need to handle the case of no-such-block must use that,
> while XLogRecGetBlockTag throws an error. The attached patch
> fixes that up, and also cleans up some random inconsistency
> about use of XLogRecHasBlockRef().
Looks reasonable.
--
Robert Haas
EDB: http://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | Nathan Bossart | 2022-04-11 21:14:35 | Re: make MaxBackends available in _PG_init |
Previous Message | Nathan Bossart | 2022-04-11 20:44:42 | Re: make MaxBackends available in _PG_init |