MIME-Version: 1.0
References: 
 <CA+HiwqE1ntHy2h9zJ9v3MwAkoGAveSERcHWkDTTZnP0kxWqbKQ@mail.gmail.com>
 <CALdSSPh=8qLa1kPsBUnPXMmigaCABiNTwmwg1wH+TmQWH=f_Ww@mail.gmail.com>
 <CA+HiwqH7TcHb=P7qHF-9jnXD3APLUWbD4QS_OJ3d-5qaC+Puzg@mail.gmail.com>
 <CA+HiwqEXyLiCdi7MQgmU8Y+aPVcOf8JpWfVA4szCmJtZAqj7Fg@mail.gmail.com>
 <6B155E97-EDE7-4CBE-8196-E709F6A4B33D@gmail.com>
In-Reply-To: <6B155E97-EDE7-4CBE-8196-E709F6A4B33D@gmail.com>
From: Amit Langote <amitlangote09@gmail.com>
Date: Thu, 16 Apr 2026 11:50:22 +0900
Message-ID: 
 <CA+HiwqFYHj70DXvY6iSYteQv6VMEiT-wk3pvETXZMM-W1bb+dg@mail.gmail.com>
Subject: Re: GetCachedPlan() refactor: move execution lock acquisition out
To: Chao Li <li.evan.chao@gmail.com>
Cc: Kirill Reshke <reshkekirill@gmail.com>,
	PostgreSQL-development <pgsql-hackers@postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: 
 <https://www.postgresql.org/message-id/CA%2BHiwqFYHj70DXvY6iSYteQv6VMEiT-wk3pvETXZMM-W1bb%2Bdg%40mail.gmail.com>
Precedence: bulk

Hi,

On Thu, Apr 16, 2026 at 11:35=E2=80=AFAM Chao Li <li.evan.chao@gmail.com> w=
rote:
> > On Apr 15, 2026, at 21:46, Amit Langote <amitlangote09@gmail.com> wrote=
:
> > On Wed, Apr 15, 2026 at 11:14=E2=80=AFAM Amit Langote <amitlangote09@gm=
ail.com> wrote:
> >> On Wed, Apr 15, 2026 at 1:08=E2=80=AFAM Kirill Reshke <reshkekirill@gm=
ail.com> wrote:
> >>>> +static bool
> >>>> +LockCachedPlan(CachedPlan *cplan)
> >>>> +{
> >>>> + AcquireExecutorLocks(cplan->stmt_list, true);
> >>>> + if (!cplan->is_valid)
> >>>> + {
> >>>> + AcquireExecutorLocks(cplan->stmt_list, false);
> >>>> + return false;
> >>>> + }
> >>>> + return true;
> >>>> +}
> >>>
> >>> simply `return cplan->is_valid ` would be more Postgres-y here, isnt =
it?
> >>
> >> Agreed, will fix too.
> >
> > Thinking more about this bit, checking only cplan->is_valid after
> > locking is not really equivalent to what CheckCachedPlan() was doing
> > before.
> >
> > CheckCachedPlan() can also reset plan->is_valid based on other state
> > (role, xmin), whereas in v1 the new check relied only on lock inval
> > callbacks to have set is_valid to false. That means the new check is
> > not enough. It may happen to be okay for the current callers in this
> > patch, but it is not something I think is safe to rely on once there
> > are future callers that do arbitrary work, such as ExecutorPrep(),
> > even if carefully coded, between the original GetCachedPlan() /
> > CheckCachedPlan() step and the later recheck after locking.
> >
> > Separately, I also realized that v1 was introducing redundant lock
> > acquisition for freshly built plans, which is not really a no behavior
> > change refactoring either.
> >
> > So in v2 I ended up reworking two parts more substantially:
> >
> > * The patch now factors the reused-generic-plan validity logic into
> > RecheckCachedPlan().
> >
> > * It also adds an is_reused output argument to GetCachedPlan(), so
> > callers can distinguish the reused-plan case from a freshly built plan
> > and avoid the extra lock step in the latter case.
>
> V2 overall looks good, it preserves the current behavior.
>
> I just have a question. RecheckCachedPlan is only called by LockCachedPla=
n and CheckCachedPlan, and both callers are static, why RecheckCachedPlan i=
s exported?

Oh, I thought someone would ask that. Yes, it could stay static in
plancache.c for now and only be exported later, when some future
smarter cached-plan locker outside plancache.c actually needs it.


--
Thanks, Amit Langote