From mboxrd@z Thu Jan  1 00:00:00 1970
From: arnd@arndb.de (Arnd Bergmann)
Date: Thu, 08 Sep 2016 22:05:13 +0200
Subject: [RFC 1/7] Make Hi3620 explicit, remove wildcards
In-Reply-To: <20160908172346.27506-1-netz.kernel@gmail.com>
References: <20160908172346.27506-1-netz.kernel@gmail.com>
Message-ID: <4234280.g2uciTKcvR@wuerfel>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Thursday, September 8, 2016 12:23:40 PM CEST Marty Plummer wrote:

>  arch/arm/Kconfig.debug | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
> index a9693b6..9094ca6 100644
> --- a/arch/arm/Kconfig.debug
> +++ b/arch/arm/Kconfig.debug
> @@ -280,7 +280,7 @@ choice
>  
>  	config DEBUG_HI3620_UART
>  		bool "Hisilicon HI3620 Debug UART"
> -		depends on ARCH_HI3xxx
> +		depends on ARCH_HI3620
>  		select DEBUG_UART_PL01X
>  		help
>  		  Say Y here if you want kernel low-level debugging support
> 

While patches should normally be split up into per-subsystem
changes, they also have to be done in an 'atomic' way:
applying just the first patch without the second one must
not introduce a regression.

In this case, I'd suggest simply doing a larger patch for
the global rename to collect 'Acked-by's and send that through
the arm-soc tree.

If you get in a similar situation with more complex changes,
you should come up with a way to do it independently. Here
you would first have to introduce a CONFIG_ARCH_HI3620
symbol in one patch and make that always selected at
the same time as CONFIG_ARCH_HI3xxx, then change all users
of that symbol, and finally remove the original.

	Arnd