linux-kernel - Re: [PATCH 3/3] printk/console: Always have a preferred console

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 for Android: free password hash cracker in your pocket

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

Date:   Wed, 14 Jun 2017 18:11:28 +0900
From:   Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
To:     Petr Mladek <pmladek@...e.com>
Cc:     Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Aleksey Makarov <aleksey.makarov@...aro.org>,
        Sabrina Dubroca <sd@...asysnail.net>,
        Sudeep Holla <sudeep.holla@....com>,
        linux-kernel@...r.kernel.org,
        Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
Subject: Re: [PATCH 3/3] printk/console: Always have a preferred console

On (06/13/17 14:54), Petr Mladek wrote:
> More consoles can be registered but one might be special. It is the one
> with CON_CONSDEV flag set. It must be the first in the console_drivers
> list. It is also sometimes called as a preferred one.
> 
> It is the console that is associated with /dev/console. It is shown
> by "showconsole" binary. But if none of the consoles have CON_CONSDEV
> flag set, the state is unclear and "showconsole" is unable to find it:
> 
>      showconsole: real console unknown: Success
> 
> Documentation/admin-guide/serial-console.rst says that the preferred
> console is the last one on the command line. But there already exists
> some fallbacks.
> 
> First, there is a fallback code that tries to enable any boot
> console and one real console if no consoles are configured.
> This code always sets CON_CONSDEV if console setup succeeded.
> 
> Second, console_unregister() sets the flag CON_CONSDEV for the next
> console in the list when a console with this flag is being removed.
> 
> Now, the flag is not set if some consoles are configured and
> the preferred one is never registered from some reason.

haha, nice. that's exactly what I was talking about a moment ago.

> This patch modifies the code that enables the configured consoles.
> It sets the CON_CONSDEV flag also when we register the first
> console. It causes that one of the registered consoles will
> always have CON_CONSDEV flag set.

hm.... my impression was that we shouldn't set CON_CONSDEV if the
console has no ->device. but then, once again, unregister_console()
does not care and register_console() cares only in one place. so I'm
a bit in doubt.

> It might have side effects. The first registered console will be
> marked as preferred and kept first in the console_drivers list
> until the really preferred one is registered. This might change
> the order of consoles in console_drivers list. As a consequence,
> another console might be selected when the really preferred one
> is unregistered. But this should require some manual intervention.
> The order was never guarantied. Therefore it does not look
> worth the effort to keep the original order.

need to think more.

	-ss

> Signed-off-by: Petr Mladek <pmladek@...e.com>
> ---
>  kernel/printk/printk.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
> index 6e651f68bffd..76b1159f2004 100644
> --- a/kernel/printk/printk.c
> +++ b/kernel/printk/printk.c
> @@ -2483,7 +2483,7 @@ void register_console(struct console *newcon)
>  		}
>  
>  		newcon->flags |= CON_ENABLED;
> -		if (i == preferred_console)
> +		if (i == preferred_console || !console_drivers)
>  			newcon->flags |= CON_CONSDEV;
>  		break;
>  	}