linux-kernel - Re: [PATCH V3 02/10] init.h: Fix the __setup

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

Date:   Wed, 1 Jul 2020 11:53:25 +0200
From:   Arnd Bergmann <arnd@...db.de>
To:     Anson Huang <anson.huang@....com>
Cc:     Russell King - ARM Linux <linux@...linux.org.uk>,
        Shawn Guo <shawnguo@...nel.org>,
        Sascha Hauer <s.hauer@...gutronix.de>,
        Sascha Hauer <kernel@...gutronix.de>,
        Fabio Estevam <festevam@...il.com>,
        Michael Turquette <mturquette@...libre.com>,
        Stephen Boyd <sboyd@...nel.org>,
        "oleksandr.suvorov@...adex.com" <oleksandr.suvorov@...adex.com>,
        Stefan Agner <stefan.agner@...adex.com>,
        Peng Fan <peng.fan@....com>, Abel Vesa <abel.vesa@....com>,
        Aisheng Dong <aisheng.dong@....com>,
        Andy Duan <fugang.duan@....com>,
        Daniel Baluta <daniel.baluta@....com>,
        YueHaibing <yuehaibing@...wei.com>,
        Stephen Rothwell <sfr@...b.auug.org.au>,
        Al Viro <viro@...iv.linux.org.uk>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        linux-clk <linux-clk@...r.kernel.org>,
        dl-linux-imx <linux-imx@....com>
Subject: Re: [PATCH V3 02/10] init.h: Fix the __setup_param() macro for module build

On Wed, Jul 1, 2020 at 11:27 AM Anson Huang <anson.huang@....com> wrote:
> > Subject: Re: [PATCH V3 02/10] init.h: Fix the __setup_param() macro for
> > module build
> >
> > On Wed, Jul 1, 2020 at 7:14 AM Anson Huang <anson.huang@....com>
> > wrote:
> >
> > I don't understand what your plan is here. Do you mean you will leave that
> > part of the clk driver as built-in?
>
> I meant I will leave the #else block of __setup_param() defined as nothing as below to
> make module build passed.
>
> #define __setup_param(str, unique_id, fn, early)        /* nothing */

No, I think that is  mistake. It will mean that other drivers with the same
bug as the imx-clk driver will appear to build fine, but not work correctly.

A build error is better than silently dropping the command line parsing in
my opinion.

> > This error just means you can't have a __setup_param() call in a loadable
> > module, which we already knew. If you need to do something with the clocks
> > early on, that has to be in built-in code and cannot be in a module. If you don't
> > need that code, then you should just remove it from both the modular version
> > and the built-in version.
> >
> > What is the purpose of that __setup_param() argument parsing in the clock
> > driver?
>
> We need the code for proper uart clock management of earlycon, from the code, it
> is trying to keep console uart clock enabled during kernel boot up.

Why not move this all to a separate file then and only build it when
CONFIG_CLK_IMX=y?
It seems that you don't need the imx_keep_uart_clocks_param() if the
clk driver is
loaded as a module, but then you also don't need the imx_clk_disable_uart()
and imx_register_uart_clocks() functions or the associated variables.

     Arnd