linux-kernel - Re: [PATCH v3] can: slcan: do not sleep with a spin lock held

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: <CAFqt6zaMnfGhwxnnJrbW1eoiGy46-WNUe8H-iqztRmZW5Z8jfQ@mail.gmail.com>
Date:   Thu, 21 Jul 2022 07:28:03 +0530
From:   Souptick Joarder <jrdr.linux@...il.com>
To:     Dario Binacchi <dario.binacchi@...rulasolutions.com>
Cc:     linux-kernel@...r.kernel.org, ltp@...ts.linux.it,
        Jeroen Hofstee <jhofstee@...tronenergy.com>,
        kbuild test robot <lkp@...el.com>,
        Richard Palethorpe <rpalethorpe@...e.de>,
        kernel test robot <oliver.sang@...el.com>, lkp@...ts.01.org,
        Jiri Slaby <jirislaby@...nel.org>,
        Linux Memory Management List <linux-mm@...ck.org>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jakub Kicinski <kuba@...nel.org>,
        Marc Kleine-Budde <mkl@...gutronix.de>,
        Paolo Abeni <pabeni@...hat.com>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Vincent Mailhol <mailhol.vincent@...adoo.fr>,
        Wolfgang Grandegger <wg@...ndegger.com>,
        linux-can@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH v3] can: slcan: do not sleep with a spin lock held

On Fri, Jul 15, 2022 at 1:00 PM Dario Binacchi
<dario.binacchi@...rulasolutions.com> wrote:
>
> We can't call close_candev() with a spin lock held, so release the lock
> before calling it. After calling close_candev(), we can update the
> fields of the private `struct can_priv' without having to acquire the
> lock.

But here we are updating private 'struct can_priv' before close_candev() while
taking the lock.  If we go by change logs, then spin_unlock_bh() need to called
before close_candev() and we can update the private 'struct can_priv'
in existing place.

>
> Fixes: c4e54b063f42f ("can: slcan: use CAN network device driver API")
> Reported-by: kernel test robot <oliver.sang@...el.com>
> Link: https://lore.kernel.org/linux-kernel/Ysrf1Yc5DaRGN1WE@xsang-OptiPlex-9020/
> Signed-off-by: Dario Binacchi <dario.binacchi@...rulasolutions.com>
>
> ---
>
> Changes in v3:
> - Update the commit message.
> - Reset sl->rcount and sl->xleft before releasing the spin lock.
>
> Changes in v2:
> - Release the lock just before calling the close_candev().
>
>  drivers/net/can/slcan/slcan-core.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/can/slcan/slcan-core.c b/drivers/net/can/slcan/slcan-core.c
> index 54d29a410ad5..d40ddc596596 100644
> --- a/drivers/net/can/slcan/slcan-core.c
> +++ b/drivers/net/can/slcan/slcan-core.c
> @@ -689,15 +689,14 @@ static int slc_close(struct net_device *dev)
>                 clear_bit(TTY_DO_WRITE_WAKEUP, &sl->tty->flags);
>         }
>         netif_stop_queue(dev);
> +       sl->rcount   = 0;
> +       sl->xleft    = 0;
> +       spin_unlock_bh(&sl->lock);
>         close_candev(dev);
>         sl->can.state = CAN_STATE_STOPPED;
>         if (sl->can.bittiming.bitrate == CAN_BITRATE_UNKNOWN)
>                 sl->can.bittiming.bitrate = CAN_BITRATE_UNSET;
>
> -       sl->rcount   = 0;
> -       sl->xleft    = 0;
> -       spin_unlock_bh(&sl->lock);
> -
>         return 0;
>  }
>
> --
> 2.32.0
>
>