linux-kernel - Re: [PATCH] platform/x86: ideapad-laptop: Fix rfkill invert

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:   Thu, 31 May 2018 15:02:53 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     Oleg Keri <ezhi99@...il.com>
Cc:     Ike Panhc <ike.pan@...onical.com>,
        Darren Hart <dvhart@...radead.org>,
        Andy Shevchenko <andy@...radead.org>,
        Platform Driver <platform-driver-x86@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] platform/x86: ideapad-laptop: Fix rfkill invert

On Thu, May 24, 2018 at 4:32 PM, Oleg Keri <ezhi99@...il.com> wrote:
> There are a lot of reports on the Internet about rfkill lock on
> modern Yoga Ideapad laptops while loading ideapad_laptop platform module.
> This patch offers the fix. Obviously it's impossible for me
> to test this patch on all ideapad laptops, so i've made an module
> parameter init_rfkill_inverted to keep old behavior.

> Comments are
> welcome.

So, it should have RFC in the Subject.

> +static bool init_rfkill_inverted;
> +module_param(init_rfkill_inverted, bool, 0444);
> +MODULE_PARM_DESC(init_rfkill_inverted, "Invert rfkill on initialization");

Module parameters are subject to not being added anymore (yes, there
are some significantly rare cases when it makes sense).

> +               /* Do not apply invert for ideapads which haven't hw switch */
> +               if (priv->has_hw_rfkill_switch || init_rfkill_inverted)
> +                       sw_blocked = !sw_blocked;

Since this is under else branch of reading value from Embedded
Controller, we need to understand what and why it returns.

For me a solution rather would be not to read EC value at all in case
of no hw switch, but I also can't test that on many possible laptops.

Would be good to get a comment from Ike.

-- 
With Best Regards,
Andy Shevchenko