Re: re. Spurious wakeup on a newly created kthread

From: Petr Mladek <pmladek@suse.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Tejun Heo <tj@kernel.org>, Lai Jiangshan <jiangshanlai@gmail.com>,
	Michal Hocko <mhocko@suse.com>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Oleg Nesterov <oleg@redhat.com>,
	"Eric W. Biederman" <ebiederm@xmission.com>
Subject: Re: re. Spurious wakeup on a newly created kthread
Date: Tue, 28 Jun 2022 11:51:37 +0200	[thread overview]
Message-ID: <YrrPKU2y4liY91u9@alley> (raw)
In-Reply-To: <CAHk-=wiC7rj1o7vTnYUPfD7YxAu09MZiZbahHqvLm9+Cgg1dFw@mail.gmail.com>

On Sat 2022-06-25 10:01:35, Linus Torvalds wrote:
> On Fri, Jun 24, 2022 at 10:00 PM Tejun Heo <tj@kernel.org> wrote:
> For example, the reason that quoted patch cannot be right is that this
> code pattern:
> 
>   while (wait_for_completion_interruptible(&worker->ready_to_start))
>     ;
> 
> is not valid kernel code. EVER. There is absolutely no way that can be correct.
> 
> Either that code can take a signal, or it cannot. If it can take a
> signal, it had better react to said signal. If it cannot, it must not
> use an interruptble sleep - since now that loop turned into a
> kernel-side busy-loop.

I agree that the code is ugly and does not make sense.

JFYI, I did it this way because I wanted to calm down the hung task
detector. kthreads ignore signals so that I did not think about
this aspect. I though that I saw this trick somewhere.

Do not ask me why I wanted to calm down the hung task detector.
I quess that I felt overloaded and wished a calm day
or something like this. I agree that it does not make
much sense.

Best Regards,
Petr