Thread: checking rd_rules in RelationBuildDesc

checking rd_rules in RelationBuildDesc

From
Ted Yu
Date:
Hi,
(In light of commit 7b2ccc5e03bf16d1e1bbabca25298108c839ec52)

In RelationBuildDesc(), we have:

        if (relation->rd_rel->relhasrules)
                RelationBuildRuleLock(relation);

I wonder if we should check relation->rd_rules after the call to RelationBuildRuleLock().

Your comment is appreciated.
Attachment

Re: checking rd_rules in RelationBuildDesc

From
Tom Lane
Date:
Ted Yu <yuzhihong@gmail.com> writes:
> I wonder if we should check relation->rd_rules after the call
> to RelationBuildRuleLock().

That patch is both pointless and wrong.  There is some
value in updating relhasrules in the catalog, so that future
relcache loads don't uselessly call RelationBuildRuleLock;
but we certainly can't try to do so right there.  That being
the case, making the relcache be out of sync with what's on
disk cannot have any good consequences.  The most likely
effect is that it would block later logic from fixing things
correctly.  There is logic in VACUUM to clean out obsolete
relhasrules flags (see vac_update_relstats), but I suspect
that would no longer work properly if we did this.

            regards, tom lane



Re: checking rd_rules in RelationBuildDesc

From
Ted Yu
Date:


On Fri, Nov 25, 2022 at 8:17 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Ted Yu <yuzhihong@gmail.com> writes:
> I wonder if we should check relation->rd_rules after the call
> to RelationBuildRuleLock().

That patch is both pointless and wrong.  There is some
value in updating relhasrules in the catalog, so that future
relcache loads don't uselessly call RelationBuildRuleLock;
but we certainly can't try to do so right there.  That being
the case, making the relcache be out of sync with what's on
disk cannot have any good consequences.  The most likely
effect is that it would block later logic from fixing things
correctly.  There is logic in VACUUM to clean out obsolete
relhasrules flags (see vac_update_relstats), but I suspect
that would no longer work properly if we did this.

                        regards, tom lane
Hi,
Thanks for evaluating the patch.

The change was originating from what we have in RelationCacheInitializePhase3():

                if (relation->rd_rel->relhasrules && relation->rd_rules == NULL)
                {
                        RelationBuildRuleLock(relation);
                        if (relation->rd_rules == NULL)
                                relation->rd_rel->relhasrules = false;

FYI