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;
MIME-Version: 1.0
References: <20201226025117.2770-1-jiangshanlai@gmail.com> <20201226025117.2770-9-jiangshanlai@gmail.com>
 <20201229100639.2086-1-hdanton@sina.com>
In-Reply-To: <20201229100639.2086-1-hdanton@sina.com>
From:   Lai Jiangshan <jiangshanlai@gmail.com>
Date:   Tue, 29 Dec 2020 18:13:03 +0800
Message-ID: <CAJhGHyCvnj1w=yO45_HxD8dgy=Xa1vUw9wsnJKtfDX2Jkm5FsQ@mail.gmail.com>
Subject: Re: [PATCH -tip V3 8/8] workqueue: Fix affinity of kworkers when
 attaching into pool
To:     Hillf Danton <hdanton@sina.com>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Valentin Schneider <valentin.schneider@arm.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Qian Cai <cai@redhat.com>,
        Vincent Donnefort <vincent.donnefort@arm.com>,
        Dexuan Cui <decui@microsoft.com>,
        Lai Jiangshan <laijs@linux.alibaba.com>,
        Tejun Heo <tj@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Dec 29, 2020 at 6:06 PM Hillf Danton <hdanton@sina.com> wrote:
>
> On Sat, 26 Dec 2020 10:51:16 +0800
> > From: Lai Jiangshan <laijs@linux.alibaba.com>
> >
> > When worker_attach_to_pool() is called, we should not put the workers
> > to pool->attrs->cpumask when there is not CPU online in it.
> >
> > We have to use wq_online_cpumask in worker_attach_to_pool() to check
> > if pool->attrs->cpumask is valid rather than cpu_online_mask or
> > cpu_active_mask due to gaps between stages in cpu hot[un]plug.
>
> In 5/8 pool->attrs->cpumask is not restored to avoid triggering
> the warning added in e9d867a67fd03ccc ("sched: Allow
> per-cpu kernel threads to run on online && !active"), is it likely
> needed to repeat that trick here?
> Is the above gap no longer existing here at the presence of
> wq_online_cpumask?

It still exists. When online, wq_online_cpumask is always
cpu_online_mask, no thing changed.

An alternative way is to move the code into a work item, which adds
the proper protection against cpu hotlug and does the work.

I don't want to add too much complex in this patchset.