From:   Thomas Gleixner <tglx@linutronix.de>
To:     Wander Lairson Costa <wander@redhat.com>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
        Waiman Long <longman@redhat.com>,
        Boqun Feng <boqun.feng@gmail.com>,
        "open list:LOCKING PRIMITIVES" <linux-kernel@vger.kernel.org>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Subject: Re: [PATCH] rtmutex: ensure we wake up the top waiter
In-Reply-To: <20230131175350.s7eiz55fozlhaegh@fedora>
References: <20230117172649.52465-1-wander@redhat.com> <875yd4k8qd.ffs@tglx>
 <CAAq0SUkN38V00HqV3Hk3ee_-=vfkKxG9xtR3n=4gAT+zCs+=Zg@mail.gmail.com>
 <87fsc6i6ud.ffs@tglx>
 <CAAq0SU=FmkSyS=-SQJBoHYEtZExK3Qn9Wqcn-c+BnmfVeO3q6g@mail.gmail.com>
 <20230131175350.s7eiz55fozlhaegh@fedora>
Date:   Thu, 02 Feb 2023 08:48:48 +0100
Message-ID: <87k010zevz.ffs@tglx>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On Tue, Jan 31 2023 at 14:53, Wander Lairson Costa wrote:
> On Tue, Jan 31, 2023 at 02:46:19PM -0300, Wander Lairson Costa wrote:
>> > If you have traces which show the sequence of lock events leading to
>> > this problem, then you should be able to decode the scenario. If you
>> > fail to extract the information, then please provide the traces so we
>> > can stare at them.
>> >
>> 
>> Here we go:
>> 
>> Let L1 and L2 be two spinlocks.
>> 
>> Let T1 be a task holding L1 and blocked on L2. T1, currently, is the top
>> waiter of L2.
>> 
>> Let T2 be the task holding L2.
>> 
>> Let T3 be a task trying to acquire L1.
>> 
>> The following events will lead to a state in which the wait queue of L2
>> isn't empty but nobody holds it.

That explains it nicely. Care to resend with proper explanations in the
changelog?

Thanks,

        tglx