Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
From:   Thomas Gleixner <tglx@linutronix.de>
To:     Anna-Maria Behnsen <anna-maria@linutronix.de>,
        linux-kernel@vger.kernel.org
Cc:     Peter Zijlstra <peterz@infradead.org>,
        John Stultz <jstultz@google.com>,
        Eric Dumazet <edumazet@google.com>,
        "Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
        Arjan van de Ven <arjan@infradead.org>,
        "Paul E . McKenney" <paulmck@kernel.org>,
        Frederic Weisbecker <frederic@kernel.org>,
        Rik van Riel <riel@surriel.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Sebastian Siewior <bigeasy@linutronix.de>,
        Giovanni Gherdovich <ggherdovich@suse.cz>,
        Lukasz Luba <lukasz.luba@arm.com>,
        "Gautham R . Shenoy" <gautham.shenoy@amd.com>,
        Srinivas Pandruvada <srinivas.pandruvada@intel.com>,
        K Prateek Nayak <kprateek.nayak@amd.com>,
        Anna-Maria Behnsen <anna-maria@linutronix.de>
Subject: Re: [PATCH v8 09/25] timer: Split out get next timer functionality
In-Reply-To: <20231004123454.15691-10-anna-maria@linutronix.de>
References: <20231004123454.15691-1-anna-maria@linutronix.de>
 <20231004123454.15691-10-anna-maria@linutronix.de>
Date:   Tue, 10 Oct 2023 00:17:09 +0200
Message-ID: <875y3fphju.ffs@tglx>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On Wed, Oct 04 2023 at 14:34, Anna-Maria Behnsen wrote:
> Split out get next timer functionality to make it reusable in other
> places. Thereby the order of getting the next expiry, forwarding the base
> clock and mark timer bases as idle, is changed. This change of order
> shouldn't have any impact, as nothing inside the function relies on the
> idle value or the updated timer base clock.

This interestingly corrects the issue of patch 8/25 because

> +	nextevt = __get_next_timer_interrupt(basej, base);
>  
>  	/*
>  	 * We have a fresh next event. Check whether we can forward the
> @@ -1952,14 +1968,10 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
>  	 */
>  	base->is_idle = time_after(nextevt, basej + 1);

base->is_idle is now handled in the correct order vs. the nextevt evaluation.
  
> -	if (base->timers_pending) {
> -		/* If we missed a tick already, force 0 delta */
> -		if (time_before(nextevt, basej))
> -			nextevt = basej;
> -		expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
> -	}
>  	raw_spin_unlock(&base->lock);

Thanks,

        tglx