Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 14 Oct 2019 14:56:32 +0200 (CEST)
From:   Thomas Gleixner <tglx@linutronix.de>
To:     Benjamin Gaignard <benjamin.gaignard@st.com>
cc:     fweisbec@gmail.com, mingo@kernel.org, marc.zyngier@arm.com,
        daniel.lezcano@linaro.org, linux-kernel@vger.kernel.org,
        linux-stm32@st-md-mailman.stormreply.com
Subject: Re: [PATCH] tick: check if broadcast device could really be
 stopped
In-Reply-To: <20191009160246.17898-1-benjamin.gaignard@st.com>
Message-ID: <alpine.DEB.2.21.1910141441350.2531@nanos.tec.linutronix.de>
References: <20191009160246.17898-1-benjamin.gaignard@st.com>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, 9 Oct 2019, Benjamin Gaignard wrote:
> @@ -78,7 +78,7 @@ static bool tick_check_broadcast_device(struct clock_event_device *curdev,
>  {
>  	if ((newdev->features & CLOCK_EVT_FEAT_DUMMY) ||
>  	    (newdev->features & CLOCK_EVT_FEAT_PERCPU) ||
> -	    (newdev->features & CLOCK_EVT_FEAT_C3STOP))
> +	    tick_broadcast_could_stop(newdev))

No. This might be called _before_ a cpuidle driver is available and then
when that driver is loaded and goes deep, everything goes south.

Aside of that it definitely breaks everything which does not use the
cpuidle stuff, which includes all machines affected by X86_BUG_AMD_APIC_C1E
and everything which uses the INTEL_IDLE driver.

Pretty much the same problem for all other places you changed.

Thanks,

	tglx