Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Valentin Schneider <valentin.schneider@arm.com>
To:     Frederic Weisbecker <frederic@kernel.org>
Cc:     "Paul E . McKenney" <paulmck@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Peter Zijlstra <peterz@infradead.org>,
        Uladzislau Rezki <urezki@gmail.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Boqun Feng <boqun.feng@gmail.com>,
        Neeraj Upadhyay <neeraju@codeaurora.org>,
        Josh Triplett <josh@joshtriplett.org>,
        Joel Fernandes <joel@joelfernandes.org>, rcu@vger.kernel.org
Subject: Re: [PATCH 10/11] rcu: Apply callbacks processing time limit only on softirq
In-Reply-To: <20211004134748.GD273854@lothringen>
References: <20210929221012.228270-1-frederic@kernel.org> <20210929221012.228270-11-frederic@kernel.org> <874ka0my57.mognet@arm.com> <20211004134748.GD273854@lothringen>
Date:   Wed, 06 Oct 2021 16:12:12 +0100
Message-ID: <87ilyakx0z.mognet@arm.com>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On 04/10/21 15:47, Frederic Weisbecker wrote:
> On Fri, Oct 01, 2021 at 06:51:32PM +0100, Valentin Schneider wrote:
>> On 30/09/21 00:10, Frederic Weisbecker wrote:
>> > Time limit only makes sense when callbacks are serviced in softirq mode
>> > because:
>> >
>> > _ In case we need to get back to the scheduler,
>> >   cond_resched_tasks_rcu_qs() is called after each callback.
>> >
>> > _ In case some other softirq vector needs the CPU, the call to
>> >   local_bh_enable() before cond_resched_tasks_rcu_qs() takes care about
>> >   them via a call to do_softirq().
>> >
>> > _ The time spent on other tasks after scheduling out, or on softirqs
>> >   processing, is spuriously accounted to the time limit.
>> >
>>
>> That wasn't the case before ("rcu: Fix callbacks processing time limit
>> retaining cond_resched()")
>
> But if cond_resched_tasks_rcu_qs() was called and then on the next iteration
> tlimit is checked, the time spent scheduling out is included, right?
>

if tlimit was set, then that branch would either continue or break; both
cases would have skipped over the cond_resched_tasks_rcu_qs() (which the
aforementioned patch addresses).