Date:   Tue, 14 Mar 2023 12:40:18 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Frederic Weisbecker <frederic@kernel.org>
Cc:     Guenter Roeck <linux@roeck-us.net>,
        Ard Biesheuvel <ardb@kernel.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        "Paul E. McKenney" <paulmck@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: Linux 6.3-rc2
Message-ID: <20230314114018.GF1845660@hirez.programming.kicks-ass.net>
References: <CAHk-=wii6BZtVKYfvQCQqbE3+t1_yAb-ea80-3PcJ4KxgpfHkA@mail.gmail.com>
 <d915df60-d06b-47d4-8b47-8aa1bbc2aac7@roeck-us.net>
 <CAHk-=wi8xQw6eTRncbJfWWYHHi0UBG2B_AfHrzZbPr=k0V_WYQ@mail.gmail.com>
 <ZA+ttVOPBvNOYHoC@lothringen>
 <dc031bc7-8ca4-7ca7-c24f-9db65b5547e4@roeck-us.net>
 <ZBBYCSZUJOWBg1s8@localhost.localdomain>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ZBBYCSZUJOWBg1s8@localhost.localdomain>
Precedence: bulk

On Tue, Mar 14, 2023 at 12:18:33PM +0100, Frederic Weisbecker wrote:
> How does the following look like as a culprit?
> 
> 62b95a7b44d1 (ARM: 9282/1: vfp: Manipulate task VFP state with softirqs disabled)

Ooh yeah, that asm implementation of local_bh_{dis,en}able completely
miss out on the lockdep state tracking. Also I think it breaks RCU, note
how __local_bh_disable_ip() explicitly does rcu_read_lock() for the
first bh-disable.