linux-kernel - Re: [PATCH v9 4/4] serial: 8250_dw: assert port->lock is held in dw8250_force

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] [day] [month] [year] [list]

Message-ID: <844iwak3u5.fsf@jogness.linutronix.de>
Date: Fri, 20 Jun 2025 14:20:10 +0206
From: John Ogness <john.ogness@...utronix.de>
To: Yunhui Cui <cuiyunhui@...edance.com>, arnd@...db.de,
 andriy.shevchenko@...ux.intel.com, benjamin.larsson@...exis.eu,
 cuiyunhui@...edance.com, gregkh@...uxfoundation.org,
 heikki.krogerus@...ux.intel.com, ilpo.jarvinen@...ux.intel.com,
 jirislaby@...nel.org, jkeeping@...usicbrands.com,
 linux-kernel@...r.kernel.org, linux-serial@...r.kernel.org,
 markus.mayer@...aro.org, matt.porter@...aro.org, namcao@...utronix.de,
 paulmck@...nel.org, pmladek@...e.com, schnelle@...ux.ibm.com,
 sunilvl@...tanamicro.com, tim.kryger@...aro.org
Subject: Re: [PATCH v9 4/4] serial: 8250_dw: assert port->lock is held in
 dw8250_force_idle()

On 2025-06-10, Yunhui Cui <cuiyunhui@...edance.com> wrote:
> Reading UART_RX and checking whether UART_LSR_DR is set should be
> atomic. Ensure the caller of dw8250_force_idle() holds port->lock.
>
> Signed-off-by: Yunhui Cui <cuiyunhui@...edance.com>
> ---
>  drivers/tty/serial/8250/8250_dw.c | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c
> index 082b7fcf251db..686f9117a3339 100644
> --- a/drivers/tty/serial/8250/8250_dw.c
> +++ b/drivers/tty/serial/8250/8250_dw.c
> @@ -13,6 +13,7 @@
>  #include <linux/delay.h>
>  #include <linux/device.h>
>  #include <linux/io.h>
> +#include <linux/lockdep.h>
>  #include <linux/mod_devicetable.h>
>  #include <linux/module.h>
>  #include <linux/notifier.h>
> @@ -117,6 +118,9 @@ static void dw8250_force_idle(struct uart_port *p)
>  	struct uart_8250_port *up = up_to_u8250p(p);
>  	unsigned int lsr;
>  
> +	/* Reading UART_LSR and UART_RX should be atomic. */
> +	lockdep_assert_held_once(&p->lock);
> +

It may be possible that during panic the port lock might not be held for
console printing:

serial8250_console_write()
  oops_in_progress and failed trylock
  serial8250_console_restore()
    serial_port_out(..., UART_LCR, ...)
      dw8250_serial_out*()
        dw8250_check_lcr()
          dw8250_force_idle()

A similar incident was discussed before [0]. In that case the result was
that the lockdep assertion was removed.

John Ogness

[0] https://lore.kernel.org/r/20230811064340.13400-1-jirislaby@kernel.org