Date:   Wed, 17 Nov 2021 18:03:16 +0100
From:   Frederic Weisbecker <frederic@kernel.org>
To:     Zqiang <qiang.zhang1211@gmail.com>
Cc:     pmladek@suse.com, akpm@linux-foundation.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH V2] watchdog: ignore nohz_full cores in new cpumask
Message-ID: <20211117170316.GA364234@lothringen>
References: <20211116061536.16293-1-qiang.zhang1211@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20211116061536.16293-1-qiang.zhang1211@gmail.com>
Precedence: bulk

On Tue, Nov 16, 2021 at 02:15:36PM +0800, Zqiang wrote:
> If the nohz_full is enabled, when update watchdog_mask, the
> nohz_full cores should be ignored.
> 
> Signed-off-by: Zqiang <qiang.zhang1211@gmail.com>

I don't know, I guess it can still be useful to have the lockup detector
available to debug some issues in nohz_full.

Those who override the cpumask should know what they are doing,
but you can still print a warning when a nohz_full CPU is included in
the range.

Thanks.

> ---
>  v1->v2:
>  if watchdog_cpumask became empty, set housekeeping_cpumask.
> 
>  kernel/watchdog.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/watchdog.c b/kernel/watchdog.c
> index ad912511a0c0..6f0c5528b399 100644
> --- a/kernel/watchdog.c
> +++ b/kernel/watchdog.c
> @@ -628,7 +628,9 @@ void lockup_detector_soft_poweroff(void)
>  static void proc_watchdog_update(void)
>  {
>  	/* Remove impossible cpus to keep sysctl output clean. */
> -	cpumask_and(&watchdog_cpumask, &watchdog_cpumask, cpu_possible_mask);
> +	cpumask_and(&watchdog_cpumask, &watchdog_cpumask, housekeeping_cpumask(HK_FLAG_TIMER));
> +	if (cpumask_empty(&watchdog_cpumask))
> +		cpumask_copy(&watchdog_cpumask, housekeeping_cpumask(HK_FLAG_TIMER));
>  	lockup_detector_reconfigure();
>  }
>  
> -- 
> 2.17.1
>