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:   Thomas Gleixner <tglx@linutronix.de>
To:     Marcelo Tosatti <mtosatti@redhat.com>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Anna-Maria Behnsen <anna-maria@linutronix.de>,
        Peter Zijlstra <peterz@infradead.org>,
        Frederic Weisbecker <frederic@kernel.org>,
        Peter Xu <peterx@redhat.com>,
        Nitesh Narayan Lal <nitesh@redhat.com>,
        Alex Belits <abelits@marvell.com>,
        "Rafael J. Wysocki" <rjw@rjwysocki.net>,
        John Stultz <john.stultz@linaro.org>
Subject: Re: [patch 8/8] hrtimer: Avoid more SMP function calls in clock_was_set()
In-Reply-To: <20210427151125.GA171315@fuller.cnet>
References: <20210427082537.611978720@linutronix.de> <20210427083724.840364566@linutronix.de> <20210427151125.GA171315@fuller.cnet>
Date:   Tue, 27 Apr 2021 21:59:43 +0200
Message-ID: <877dkno5w0.ffs@nanos.tec.linutronix.de>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On Tue, Apr 27 2021 at 12:11, Marcelo Tosatti wrote:
> On Tue, Apr 27, 2021 at 10:25:45AM +0200, Thomas Gleixner wrote:
> Consider
>
>
> hrtimer_interrupt
> in_hrtirq = 1
> __run_hrtimer
> raw_spin_unlock_irqrestore(&cpu_base->lock, flags)
> 								settimeofday
> 								clock_was_set
> 								lock cpu_base->lock
> 								update_needs_ipi returns false
> continue to process hrtimers with stale base->offset

Bah. I somehow convinced myself that hrtimer_interrupt() rechecks the
offset before clearing in_hrtirq, but that's not true. It does so when
reprogramming fails, but not for the regular case.

Lemme stare at it some more.

Thanks,

        tglx