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;
Date:   Fri, 8 Apr 2022 08:58:34 -0700
From:   "Paul E. McKenney" <paulmck@kernel.org>
To:     Steven Rostedt <rostedt@goodmis.org>
Cc:     Joel Fernandes <joel@joelfernandes.org>,
        Kalesh Singh <kaleshsingh@google.com>,
        Suren Baghdasaryan <surenb@google.com>,
        kernel-team <kernel-team@android.com>, Tejun Heo <tj@kernel.org>,
        Tim Murray <timmurray@google.com>, Wei Wang <wvw@google.com>,
        Kyle Lin <kylelin@google.com>,
        Chunwei Lu <chunweilu@google.com>,
        Lulu Wang <luluw@google.com>,
        Frederic Weisbecker <frederic@kernel.org>,
        Neeraj Upadhyay <quic_neeraju@quicinc.com>,
        Josh Triplett <josh@joshtriplett.org>,
        Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
        Lai Jiangshan <jiangshanlai@gmail.com>,
        rcu <rcu@vger.kernel.org>, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2] EXP rcu: Move expedited grace period (GP) work to RT
 kthread_worker
Message-ID: <20220408155834.GI4285@paulmck-ThinkPad-P17-Gen-1>
Reply-To: paulmck@kernel.org
References: <20220408045734.1158817-1-kaleshsingh@google.com>
 <CAEXW_YSqY2nFZrn4AjpUzJ+dwZc7jaVMG9RG5gvTyb3zFYWtQA@mail.gmail.com>
 <20220408113946.056c083e@gandalf.local.home>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220408113946.056c083e@gandalf.local.home>
Precedence: bulk

On Fri, Apr 08, 2022 at 11:39:46AM -0400, Steven Rostedt wrote:
> On Fri, 8 Apr 2022 06:21:03 -0400
> Joel Fernandes <joel@joelfernandes.org> wrote:
> 
> > > +       default !PREEMPT_RT && NR_CPUS <= 32  
> > 
> > What is the benefit of turning it off on PREEMPT_RT, even if
> > PREEMPT_RT does not use expedited GPs much post-boot? I would think in
> > the future if PREEMPT_RT ever uses expedited GPs, they would want this
> > feature even more. I'd rather be future-proof now as I don't see any
> > advantages of disabling it on !PREEMPT_RT (And a drawback that the fix
> > won't apply to those systems). Also will keep the config simple.
> 
> The default kthread priority is 1. This should not affect PREEMPT_RT, as
> PREEMPT_RT users are usually more concerned by the performance of their
> higher priority tasks. Priority 1 is not considered an issue.
> 
> I do not see why PREEMPT_RT is special here. Why was it singled out?

Because I didn't see the point of doing kthread_create_worker() to create
a pair of kthread_worker structuress that were never going to be used
in PREEMPT_RT kernels.

Or are PREEMPT_RT kernels no longer booting with rcupdate.rcu_normal?
Last I knew, they all did this to avoid IPI latencies from expedited
grace periods.  Has this changed?

							Thanx, Paul