Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Joel Fernandes <joel@joelfernandes.org>
Mime-Version: 1.0 (1.0)
Subject: Re: [PATCH RFC] rcu/kfree: Do not request RCU when not needed
Date:   Wed, 2 Nov 2022 15:51:49 -0400
Message-Id: <ED8E1017-A219-4815-BA28-5F44A2FB8FF2@joelfernandes.org>
References: <20221102183237.GO5600@paulmck-ThinkPad-P17-Gen-1>
Cc:     Uladzislau Rezki <urezki@gmail.com>, rcu@vger.kernel.org,
        linux-kernel@vger.kernel.org
In-Reply-To: <20221102183237.GO5600@paulmck-ThinkPad-P17-Gen-1>
To:     paulmck@kernel.org
Precedence: bulk


> On Nov 2, 2022, at 2:32 PM, Paul E. McKenney <paulmck@kernel.org> wrote:
>=20
> =EF=BB=BFOn Wed, Nov 02, 2022 at 06:30:24PM +0100, Uladzislau Rezki wrote:=

>>>=20
>>> to wait for a GP. Or if the poll_state_synchronize_rcu(oldstate)) !=3D 0=

>>> queue_rcu_work().
>>>=20
>> A small fix. If poll_state_synchronize_rcu(oldstate)) =3D=3D 0 then
>> queue_rcu_work() since a GP is still in progress.
>=20
> Both get_state_synchronize_rcu() and poll_state_synchronize_rcu()
> may be invoked from interrupt handlers.  And from NMI handlers,
> for that matter.

Thanks for clarifying. I meant to say cond_synchronize_rcu() cannot be calle=
d from interrupt context, but I made that comment before Vlad cooked his pat=
ch.

Thanks,

 - Joel


>                            Thanx, Paul