linux-kernel - Re: [PATCH RT v3 1/5] rcu: Acquire RCU lock when disabling BHs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20190917144252.v34ina4z2ydchoit@linutronix.de>
Date:   Tue, 17 Sep 2019 16:42:52 +0200
From:   Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To:     Scott Wood <swood@...hat.com>
Cc:     linux-rt-users@...r.kernel.org, linux-kernel@...r.kernel.org,
        "Paul E . McKenney" <paulmck@...ux.ibm.com>,
        Joel Fernandes <joel@...lfernandes.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Steven Rostedt <rostedt@...dmis.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Juri Lelli <juri.lelli@...hat.com>,
        Clark Williams <williams@...hat.com>
Subject: Re: [PATCH RT v3 1/5] rcu: Acquire RCU lock when disabling BHs

On 2019-09-17 09:06:28 [-0500], Scott Wood wrote:
> Sorry, I missed that you were asking about rcu_read_lock_bh() as well.  I
> did remove the change to rcu_read_lock_bh_held().

Sorry for not being clear here.

> With this patch, local_bh_disable() calls rcu_read_lock() on RT which
> handles this debug stuff.  Doing it twice shouldn't be explicitly harmful,
> but it's redundant, and debug kernels are slow enough as is.

rcu_read_lock() does:
|         __rcu_read_lock();
|         __acquire(RCU);
|         rcu_lock_acquire(&rcu_lock_map);
|         RCU_LOCKDEP_WARN(!rcu_is_watching(),
|                          "rcu_read_lock() used illegally while idle");

__acquire() is removed removed by cpp.
That RCU_LOCKDEP_WARN is doing the same thing as above and redundant.
Am I right to assume that you consider
	rcu_lock_acquire(&rcu_bh_lock_map);

redundant because the only user of that is also checking for
rcu_lock_map?

> -Scott

Sebastian