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:   Mon, 7 Feb 2022 11:48:45 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     YT Chang <yt.chang@mediatek.com>
Cc:     Ingo Molnar <mingo@redhat.com>, Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
        linux-mediatek@lists.infradead.org, wsd_upstream@mediatek.com
Subject: Re: [PATCH 1/1] sched: Add update_rq_clock() in sched_rt_rq_enqueue()
Message-ID: <20220207104845.GD23216@worktop.programming.kicks-ass.net>
References: <20211224012239.4694-1-yt.chang@mediatek.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20211224012239.4694-1-yt.chang@mediatek.com>
Precedence: bulk

On Fri, Dec 24, 2021 at 09:22:39AM +0800, YT Chang wrote:
> Add update_rq_clock() in sched_rt_rq_enqueue() to
> prevent the warning "rq->clock_update_flags < RQCF_ACT_SKIP"
> when call rq_clock() in cpufreq_update_util().
> 
> sched_rt_rq_enqueue ->
>    enqueue_top_rt_rq ->
>       cpufreq_update_util ->
>          rq_clock ->
>             assert_clock_updated
> 
> Signed-off-by: YT Chang <yt.chang@mediatek.com>
> Change-Id: I4fba5a561b7064aafa991d7f1a34431607779cb4

Change-Id does not belong in kernel patches..

> ---
>  kernel/sched/rt.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> index b48baaba2fc2..faf1a68c0723 100644
> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -539,6 +539,7 @@ static void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
>  
>  	int cpu = cpu_of(rq);
>  
> +	update_rq_clock(rq);

This isn't right I think; there's at least one caller of this function
that already did that. A double clock update is also wrong. Each path
should have just one.

>  	rt_se = rt_rq->tg->rt_se[cpu];
>  
>  	if (rt_rq->rt_nr_running) {
> -- 
> 2.18.0
>