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;
Date:   Wed, 14 Dec 2022 13:03:59 +0200
From:   Vladimir Oltean <olteanv@gmail.com>
To:     Maksim Kiselev <bigunclemax@gmail.com>
Cc:     fido_max@inbox.ru, mw@semihalf.com, Andrew Lunn <andrew@lunn.ch>,
        Florian Fainelli <f.fainelli@gmail.com>,
        "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        Richard Cochran <richardcochran@gmail.com>,
        Russell King <linux@armlinux.org.uk>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: Subject: Locking mv88e6xxx_reg_lock twice leads deadlock for
 88E6176 switch
Message-ID: <20221214110359.y4kam23sxby7vpek@skbuf>
References: <20221213225856.1506850-1-bigunclemax@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20221213225856.1506850-1-bigunclemax@gmail.com>
Precedence: bulk

On Wed, Dec 14, 2022 at 01:58:55AM +0300, Maksim Kiselev wrote:
> Hello, friends.

Hello.

> I have a device with Marvell 88E6176 switch. 
> After 'mv88e6xxx: fix speed setting for CPU/DSA ports (cc1049ccee20)'commit was applied to
> mainline kernel I faced with a problem that switch driver stuck at 'mv88e6xxx_probe' function.

Sorry for that.

> I made some investigations and found that 'mv88e6xxx_reg_lock' called twice from the same thread which leads to deadlock.
> 
> I added logs to 'mv88e6xxx_reg_lock' and 'mv88e6xxx_reg_unlock' functions to see what happened.

I hope you didn't spend too much time doing that. If you enable CONFIG_PROVE_LOCKING,
you should automatically get a stack trace with the two threads that
acquired the mutex leading to the deadlock.

I've sent a patch which solves that issue here:
https://patchwork.kernel.org/project/netdevbpf/patch/20221214110120.3368472-1-vladimir.oltean@nxp.com/
I've regression-tested it on 88E6390. Please confirm with a Tested-by:
tag on that patch that it does resolve the deadlock for 88E6176.

Thanks for reporting!