Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 26 Jun 2018 09:37:22 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Uwe =?iso-8859-1?Q?Kleine-K=F6nig?= 
        <u.kleine-koenig@pengutronix.de>
Cc:     linux-kernel@vger.kernel.org, Oleg Nesterov <oleg@redhat.com>,
        "Eric W . Biederman" <ebiederm@xmission.com>,
        "Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Gavin Schenk <g.schenk@eckelmann.de>, kernel@pengutronix.de
Subject: Re: [PATCH] RFC: siox: don't create a thread without starting it
Message-ID: <20180626073722.GX2458@hirez.programming.kicks-ass.net>
References: <20180625102056.28468-1-u.kleine-koenig@pengutronix.de>
 <20180625125105.GZ2494@hirez.programming.kicks-ass.net>
 <20180625192121.k3hx32xbbhqkyfu4@pengutronix.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20180625192121.k3hx32xbbhqkyfu4@pengutronix.de>
User-Agent: Mutt/1.10.0 (2018-05-17)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jun 25, 2018 at 09:21:21PM +0200, Uwe Kleine-K?nig wrote:
> > That said, irrespective of the whole UNINTERRUPTIBLE/IDLE thing, I find
> > this construct fairly fragile. We rely on not getting any spurious
> > wakeups without a 'special' state.
> 
> Well, if the thread is woken up unintentionally nothing happens (apart

In your case sure. But people rely on the task being 'idle/blocked'
after kthread_create() to call things like kthread_bind() on it.

If it were to be woken early because dodgy games, the kthread_bind()
would fail, which in turn can lead to all sorts of 'fun' problems.

That is the whole point of this intermediate blocked state, it allows
you to muck with the kthread before it starts running custom code.
Arguably one could do it differently, but this is what it is.