Received-SPF: pass (google.com: domain of linux-kernel+bounces-95530-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
From: Uladzislau Rezki <urezki@gmail.com>
Date: Thu, 7 Mar 2024 13:25:03 +0100
To: Joel Fernandes <joel@joelfernandes.org>
Cc: "Uladzislau Rezki (Sony)" <urezki@gmail.com>,
	"Paul E . McKenney" <paulmck@kernel.org>, RCU <rcu@vger.kernel.org>,
	Neeraj upadhyay <Neeraj.Upadhyay@amd.com>,
	Boqun Feng <boqun.feng@gmail.com>, Hillf Danton <hdanton@sina.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Oleksiy Avramchenko <oleksiy.avramchenko@sony.com>,
	Frederic Weisbecker <frederic@kernel.org>
Subject: Re: [PATCH 1/2] rcu: Do not release a wait-head from a GP kthread
Message-ID: <ZemyHxqjbiUrb9kf@pc636>
References: <20240305195720.42687-1-urezki@gmail.com>
 <a1faf101-c689-4530-a9a5-c7f95b8825d6@joelfernandes.org>
 <CAEXW_YTzMPxdm=UmMScppfP2WTqoyo6Z4c0p06HNgT-NAL2ciQ@mail.gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <CAEXW_YTzMPxdm=UmMScppfP2WTqoyo6Z4c0p06HNgT-NAL2ciQ@mail.gmail.com>

On Wed, Mar 06, 2024 at 05:44:04PM -0500, Joel Fernandes wrote:
> On Wed, Mar 6, 2024 at 5:31 PM Joel Fernandes <joel@joelfernandes.org> wrote:
> >
> >
> >
> > On 3/5/2024 2:57 PM, Uladzislau Rezki (Sony) wrote:
> > > Fix a below race by not releasing a wait-head from the
> > > GP-kthread as it can lead for reusing it whereas a worker
> > > can still access it thus execute newly added callbacks too
> > > early.
> > >
> [...]
> > There might a way to prevent queuing new work as fast-path optimization, incase
> > the CBs per GP will always be < SR_MAX_USERS_WAKE_FROM_GP but I could not find a
> > workqueue API that helps there, and work_busy() has comments saying not to use that.
> 
> One way to do this would be to maintain a count of how many CBs are in
> flight via the worker route, and then fast-path-free the thing if the
> count is 0. Should I send a patch around something like that? It saves
> 1 more wakeup per synchronize_rcu() I think.
> 
We can release the last wait-head if we know that the worker is not
pending/running. Then we guarantee that Frederic's case is not possible.
From the other hand it will introduce again more mess because the idea
was, in the begging, to start with something really simple :)

--
Uladzislau Rezki