Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Feedback-ID: iad51458e:Fastmail
Date:   Fri, 16 Sep 2022 10:46:18 -0700
From:   Boqun Feng <boqun.feng@gmail.com>
To:     Waiman Long <longman@redhat.com>
Cc:     Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Shaokun Zhang <zhangshaokun@hisilicon.com>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Petr Mladek <pmladek@suse.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Ben Dooks <ben.dooks@sifive.com>,
        Rasmus Villemoes <linux@rasmusvillemoes.dk>,
        Luis Chamberlain <mcgrof@kernel.org>,
        Xiaoming Ni <nixiaoming@huawei.com>,
        John Ogness <john.ogness@linutronix.de>,
        LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH (repost)] locking/lockdep: add
 debug_show_all_lock_holders()
Message-ID: <YyS2aoPoC2TznZHO@boqun-archlinux>
References: <d5393b0e-a296-3296-d376-c9178669747b@I-love.SAKURA.ne.jp>
 <3e027453-fda4-3891-3ec3-5623f1525e56@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3e027453-fda4-3891-3ec3-5623f1525e56@redhat.com>
Precedence: bulk

On Fri, Sep 16, 2022 at 10:51:14AM -0400, Waiman Long wrote:
[...]
> > 
> > I'm expecting lockdep to continue working without calling printk() even after
> > one of problem patterns (e.g. circular locking dependency) was printk()ed, so that
> > debug_show_all_locks()/debug_show_all_lock_holders() can call printk() when needed.
> > 
> > Changing debug_locks behavior is a future patch. For now, this patch alone
> > will help debugging Greg's usb.git#usb-testing tree which is generating
> > many "INFO: task hung in" reports.

The default behavior today does make sense: if the system has one
deadlock it should be fixed first before searching out another. So I
expect that changing debug_locks behavior would be configurable and we
keep the default as today.

> 
> Boqun is actually working on a modularization patch to make some lockdep
> checking still active even after a lockdep bug is reported. I think he will
> take into consideration about this request.
> 

Yes, this is one of the "problems" I try to resolve while cleaning up
the lockdep.

Regards,
Boqun

> Cheers,
> Longman
>