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;
Date:   Mon, 22 Jun 2020 19:29:36 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     "Rafael J. Wysocki" <rafael@kernel.org>
Cc:     Chen Yu <yu.c.chen@intel.com>, Len Brown <lenb@kernel.org>,
        Daniel Lezcano <daniel.lezcano@linaro.org>,
        Ingo Molnar <mingo@kernel.org>,
        Linux PM <linux-pm@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Rui Zhang <rui.zhang@intel.com>
Subject: Re: [PATCH][v2] PM / s2idle: Clear _TIF_POLLING_NRFLAG before
 suspend to idle
Message-ID: <20200622172936.GA4781@hirez.programming.kicks-ass.net>
References: <20200616040442.21515-1-yu.c.chen@intel.com>
 <CAJZ5v0gBVBAjdCOXsM-Fa-iAkuv2JMi2mVkG5w7ADcg9dWencA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAJZ5v0gBVBAjdCOXsM-Fa-iAkuv2JMi2mVkG5w7ADcg9dWencA@mail.gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jun 22, 2020 at 06:19:35PM +0200, Rafael J. Wysocki wrote:
> > Fixes: b2a02fc43a1f ("smp: Optimize send_call_function_single_ipi()")
> > Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> > Reported-by: kbuild test robot <lkp@intel.com>
> > Cc: "Rafael J. Wysocki" <rafael@kernel.org>
> > Cc: Len Brown <lenb@kernel.org>
> > Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
> > Signed-off-by: Chen Yu <yu.c.chen@intel.com>
> 
> Peter, any more comments here?

Only that the whole s2idle stuff could do with a cleanup :-)

> > +static int call_s2idle(struct cpuidle_driver *drv, struct cpuidle_device *dev,
> > +                      int index)
> > +{
> > +       if (!current_clr_polling_and_test())
> > +               s2idle_enter(drv, dev, index);
> > +
> > +       return index;
> 
> Is the value returned here used at all?
> 
> > +}
> > +
> >  /**
> >   * cpuidle_enter_s2idle - Enter an idle state suitable for suspend-to-idle.
> >   * @drv: cpuidle driver for the given CPU.
> > @@ -187,7 +197,7 @@ int cpuidle_enter_s2idle(struct cpuidle_driver *drv, struct cpuidle_device *dev)
> >          */
> >         index = find_deepest_state(drv, dev, U64_MAX, 0, true);
> >         if (index > 0)
> > -               enter_s2idle_proper(drv, dev, index);
> > +               call_s2idle(drv, dev, index);
> 
> I'm wondering why this can't be
> 
>     if (index > 0 && !current_clr_polling_and_test())
>             enter_s2idle_proper(drv, dev, index);

Works for me. Some Wysocki guy wrote much of it, best ask him :-)

The thing that confused me is that all this is way different from the
normal idle path and didn't keep the invariants.

Ideally; much of that gets folded back into the normal patch somehow.