Received-SPF: pass (google.com: domain of linux-kernel+bounces-174193-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <a407d12f-6845-4f51-a112-6bdc17641ff1@web.de>
Date: Thu, 9 May 2024 08:28:21 +0200
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v3] time/tick-sched: idle load balancing when nohz_full
 cpu becomes idle.
To: Levi Yun <ppbuk5246@gmail.com>, kernel-janitors@vger.kernel.org,
 Anna-Maria Behnsen <anna-maria@linutronix.de>,
 Frederic Weisbecker <frederic@kernel.org>, Ingo Molnar <mingo@kernel.org>,
 Thomas Gleixner <tglx@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>
References: <20240508172621.30069-1-ppbuk5246@gmail.com>
 <20240508192213.31050-1-ppbuk5246@gmail.com>
Content-Language: en-GB
From: Markus Elfring <Markus.Elfring@web.de>
In-Reply-To: <20240508192213.31050-1-ppbuk5246@gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
UI-OutboundReport: notjunk:1;M01:P0:VQqKqKPXP9Q=;Sy19RzSEH5REP179ojt6y9bn5Af
 BTmggMFcoXwYL8dH2hFzuVMCMnDbn7MLLkAn4AEh7But3T+25jYFixSfMhySXKAvhDXx6LSsi
 tz727qrk10Yh9SrZlk4Ogy1IzfZX/fbnkW7aNIefNUWyJEF9icPkrLVTv0Ik3cjIHJR0vRN8P
 1YWrekTjWsTp2b4VxXqZkqn/OFnyhsyL1Ecm/F5kazpWcI0r5AD4gaTr1xX9VHwmMmUw1iUeg
 qv5eNYpLM12MNeIhuBaoqlS/72ODEsQO0QVONndEmide0XlBDA7Mmx3X+6UCgRQol/B+uwBZb
 2Ln/r4kYnS8aVPDBCzqd32xWVdnf6PJFQObMeh0BSF61X2Kuu8HyObfoPGdoD6042UG1+GlOw
 Xn7LyaZSUHjLAmx4s9Dn4tthOrFfe7b9CUJaxKjz1l8yW9H8D84zm/EdmY79kr4BuSQvXnZ8n
 4faw5yO0DQ8C0lUjdjAby+B4lzSwGJth1KSSk6nWcO/jb79o6T/IYkj0Syr9W3NvgAbypVIhg
 KVt+FYIWQpr3uUZDUNLQSSRWhtY0phzS2eF57fyu64pS7xb1sjVqGRO4QBx8IOoY/8Pkg1aA4
 5rH3MbucpnyH0ywlSWu7QZDTJvfF/ITJbsViZwrMhyiMfvkh3FM/SnCJV2GR2MRI3dpVBuKDP
 Xtjf+mHdMH3rHC1ZoN8LzWDdzpX87WUjz/LIsNetEcuZOh9uO8GS1RGtmRQDHSDvN7E9kQ3Zh
 ghlYDellhGOpEss9Ivp/7IdqTcIghO/KlkVROU19jgNf9Q7yM/+zggEAgQ29anYCfXpm0UDmo
 aJx6axTv+tU4PomGjVrRJeTBbseOSV7DJ/4nwILB3+cFg=

> When nohz_full CPU stops tick in tick_nohz_irq_exit(),
> It wouldn't be chosen to perform idle load balancing because it doesn't
> call nohz_balance_enter_idle() in tick_nohz_idle_stop_tick() when it
> becomes idle.
>
> tick_nohz_idle_stop_tick() is only called in idle state and
> nohz_balance_enter_idle() tracks the CPU which is part of nohz.idle_cpus=
_mask
> with rq->nohz_tick_stopped.
>
> Change tick_nohz_idle_stop_tick() to call nohz_balance_enter_idle()
> without checking !was_stopped so that nohz_full cpu can be chosen to
> perform idle load balancing when it enters idle state.

Would you eventually like to add the tag =E2=80=9CFixes=E2=80=9D once more=
?


=E2=80=A6
> +++ b/kernel/time/tick-sched.c
> @@ -1228,8 +1228,10 @@ void tick_nohz_idle_stop_tick(void)
>  		ts->idle_sleeps++;
>  		ts->idle_expires =3D expires;
>
> -		if (!was_stopped && tick_sched_flag_test(ts, TS_FLAG_STOPPED)) {
> -			ts->idle_jiffies =3D ts->last_jiffies;
> +		if (tick_sched_flag_test(ts, TS_FLAG_STOPPED)) {
> +			if (!was_stopped)
> +				ts->idle_jiffies =3D ts->last_jiffies;
> +
>  			nohz_balance_enter_idle(cpu);
>  		}
=E2=80=A6

I interpret these diff data in the way that you propose to reorder
two condition checks.

But I wonder still how =E2=80=9Cgood=E2=80=9D the presented change descrip=
tion fits to
the suggested source code adjustment.

Regards,
Markus