Re: [PATCH] fix the -test3 input config damages

[Vojtech Pavlik <vojtech@suse.cz>]

On Fri, Aug 22, 2003 at 06:38:00PM +0200, Christoph Hellwig wrote:

> There's really no point in forcing in support for all kinds of
> optional input devices unless CONFIG_EMBEDDED.  Also some of the
> selections really broke configs that worked fine before as indicated
> by the reports on lkml.  Instead select CONFIG_INPUT if CONFIG_VT
> is selected so people upgrading from 2.4 using make oldconfig have
> a chance to see it if they didn't need CONFIG_INPUT before.
> 
> Btw, could we please get a consensus on what CONFIG_EMBEDDED is
> supposed to mean?  It was introduced to allow compiling code out
> for special cases that normal userspace should be able to rely
> on like epoll and futexes but people seem to use it as an Aunt Tillie
> guard these days..

I think this is the way to go. I'm merging this into my tree now.

> --- 1.15/drivers/char/Kconfig	Wed Jul 16 13:39:32 2003
> +++ edited/drivers/char/Kconfig	Sun Aug 10 11:17:02 2003
> @@ -5,8 +5,8 @@
>  menu "Character devices"
>  
>  config VT
> -	bool "Virtual terminal" if EMBEDDED
> -	requires INPUT=y
> +	bool "Virtual terminal"
> +	select INPUT
>  	default y
>  	---help---
>  	  If you say Y here, you will get support for terminal devices with
> @@ -36,7 +36,7 @@
>  	  shiny Linux system :-)
>  
>  config VT_CONSOLE
> -	bool "Support for console on virtual terminal" if EMBEDDED
> +	bool "Support for console on virtual terminal"
>  	depends on VT
>  	default y
>  	---help---

[snip]

-- 
Vojtech Pavlik
SuSE Labs, SuSE CR