linux-mips
[Top] [All Lists]

Re: [PATCH] no console disabling during suspend stage

To: Rodolfo Giometti <giometti@linux.it>
Subject: Re: [PATCH] no console disabling during suspend stage
From: Andrew Morton <akpm@osdl.org>
Date: Fri, 21 Jul 2006 06:05:13 -0700
Cc: linux-kernel@vger.kernel.org, linux-mips@linux-mips.org
In-reply-to: <20060719091559.GD25330@enneenne.com>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20060719091559.GD25330@enneenne.com>
Sender: linux-mips-bounce@linux-mips.org
On Wed, 19 Jul 2006 11:16:00 +0200
Rodolfo Giometti <giometti@linux.it> wrote:

> here a little patch to avoid disabling console during suspend stage
> while we are debugging kernel.
> 
> ...
>
> --- kernel/power/main.c       2006-07-19 10:54:11.000000000 +0200
> +++ kernel/power/main.c.new   2006-07-18 19:38:41.000000000 +0200
> @@ -86,7 +86,9 @@
>                       goto Thaw;
>       }
>  
> +#ifndef CONFIG_DEBUG_KERNEL
>       suspend_console();
> +#endif

CONFIG_DEBUG_KERNEL is not really the appropriate thing to use here - it's
more a user-interface/Kconfig level thing.

Can you please describe the problems which suspend_console() are causing?

Some runtime knob might be more appropriate.  Perhaps /sys/power/debug?

If the suspend_console() call is to be disabled then you'll want to disable
the matching resume_console() call too.

printk.c needs sem2mutex conversion.

The console_suspended/secondary_console_sem handling is racy.  Bad Linus. 
But as we're running on a single CPU and all other tasks are frozen it
doesn't matter.

<Prev in Thread] Current Thread [Next in Thread>