linux-kernel - Re: [PATCH] clk: mediatek: remove superfluous line including linux/clk-provider.h

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]

Date:   Wed, 27 Dec 2017 12:01:09 +0800
From:   Sean Wang <sean.wang@...iatek.com>
To:     Stephen Boyd <sboyd@...eaurora.org>
CC:     <mturquette@...libre.com>, <matthias.bgg@...il.com>,
        <jdelvare@...e.de>, <jamesjj.liao@...iatek.com>,
        <weiyi.lu@...iatek.com>, <kevin-cw.chen@...iatek.com>,
        <shunli.wang@...iatek.com>, <chen.zhong@...iatek.com>,
        <arnd@...db.de>, <linux-mediatek@...ts.infradead.org>,
        <linux-clk@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] clk: mediatek: remove superfluous line including
 linux/clk-provider.h

On Tue, 2017-12-26 at 17:15 -0800, Stephen Boyd wrote:
> On 12/25, sean.wang@...iatek.com wrote:
> > From: Sean Wang <sean.wang@...iatek.com>
> > 
> > clk-mtk.h has already included linux/clk-provider.h inside, so remove
> > the superfluous line including linux/clk-provider.h from these files which
> > also included clk-mtk.h.
> > 
> > Signed-off-by: Sean Wang <sean.wang@...iatek.com>
> > Cc: Jean Delvare <jdelvare@...e.de>
> > ---
> 
> Please don't do this. If anything, I would remove as many
> includes from clk-mtk.h as possible and forward declare the
> structures used in there so there isn't any header dependency.
> Then it becomes clear which C files are clk-provider drivers
> almost immediately because we have the include in the driver, and
> we also avoid any header dependency chains where we must include
> some header file if we want to get another header file's
> definitions.
> 

Agreed on above. 

It really becomes clearer to know which c files belong
to clock provider driver, not only just considering header dependency.
For these points, it should be kept there.

Thanks for your detailed explanation

	Sean