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;
MIME-Version: 1.0
References: <20190321075725.14054-1-duyuyang@gmail.com> <20190321075725.14054-19-duyuyang@gmail.com>
In-Reply-To: <20190321075725.14054-19-duyuyang@gmail.com>
From:   Yuyang Du <duyuyang@gmail.com>
Date:   Thu, 4 Apr 2019 13:03:10 +0800
Message-ID: <CAHttsrZ0JgWvwtcaR-qbV8dANhXyOkNypw=GjfPhh1P8J4T1Aw@mail.gmail.com>
Subject: Question on a lockdep test case about mixed read-write ABBA
To:     peterz@infradead.org, will.deacon@arm.com, mingo@kernel.org
Cc:     Bart Van Assche <bvanassche@acm.org>, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Peter,

I observed this test case you wrote in Commit: e9149858974606
("locking/lockdep/selftests: Add mixed read-write ABBA").

static void rwsem_ABBA2(void)
{
       RSL(X1);
       ML(Y1);
       MU(Y1);
       RSU(X1);

       ML(Y1);
       RSL(X1);
       RSU(X1);
       MU(Y1); // should fail
}

Why should it fail? This is not a deadlock, right? The depencencies
would be built by lockdep though; that results in a false positive.

Thanks,
Yuyang