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;
Date:   Thu, 8 Jul 2021 13:35:26 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Juri Lelli <juri.lelli@redhat.com>
Cc:     mingo@redhat.com, linux-kernel@vger.kernel.org,
        vincent.guittot@linaro.org, rostedt@goodmis.org,
        dietmar.eggemann@arm.com, bristot@redhat.com, bsegall@google.com,
        mgorman@suse.de, Mark Simmons <msimmons@redhat.com>
Subject: Re: [PATCH] sched/rt: Fix double enqueue caused by rt_effective_prio
Message-ID: <YObi/nNu115qg5Rr@hirez.programming.kicks-ass.net>
References: <20210701091431.256457-1-juri.lelli@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210701091431.256457-1-juri.lelli@redhat.com>
Precedence: bulk

On Thu, Jul 01, 2021 at 11:14:31AM +0200, Juri Lelli wrote:
> Looks like this survives the test case (and regression testing), but I'm
> still not fully convinced that the above statement "Concurrent priority
> inheritance handling is still safe and will eventually converge to a new
> state by following the inheritance chain(s)" is actually sound.

I think we good there. rt_mutex_adjust_prio_chain() updates ->prio in
step 7 and calls rt_mutex_setprio() in step 11.

So if we race against __sched_setschedule() and observe the old value,
step 11 will follow up and correct us.