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;
MIME-Version: 1.0
References: <CANN689HMLC4uKCjztqfZsDO7pPbsvViOQQ_HDWWYU4=8cZbKvQ@mail.gmail.com>
In-Reply-To: <CANN689HMLC4uKCjztqfZsDO7pPbsvViOQQ_HDWWYU4=8cZbKvQ@mail.gmail.com>
From:   Michel Lespinasse <walken@google.com>
Date:   Sat, 22 Aug 2020 10:30:34 -0700
Message-ID: <CANN689EscUK7_ZaiUbCdt5K-=PYPFgVdeE_GFem0A7WZVfQH-A@mail.gmail.com>
Subject: Re: Lockdep question regarding two-level locks
To:     Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
        Davidlohr Bueso <dave@stgolabs.net>
Cc:     LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Sat, Aug 22, 2020 at 9:04 AM Michel Lespinasse <walken@google.com> wrote:
> - B's implementation could, when lockdep is enabled, always release
> lock A before acquiring lock B. This is not ideal though, since this
> would hinder testing of the not-blocked code path in the acquire
> sequence.

Actually, this may be an acceptable way to handle my issue. In the
non-blocking case, B's implementation does not have to actually
release A, but it could tell lockdep that it's released A, acquired B
and acquired A again. Kinda ugly but should work...

-- 
Michel "Walken" Lespinasse
A program is never fully debugged until the last user dies.