linux-kernel - Re: [PATCH] pinctrl/rockchip: fix gpio device creation

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, 2 Dec 2021 02:25:27 +0100
From:   Linus Walleij <linus.walleij@...aro.org>
To:     John Keeping <john@...anate.com>
Cc:     linux-gpio@...r.kernel.org, Heiko Stuebner <heiko@...ech.de>,
        Jianqun Xu <jay.xu@...k-chips.com>,
        linux-arm-kernel@...ts.infradead.org,
        linux-rockchip@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] pinctrl/rockchip: fix gpio device creation

On Fri, Nov 26, 2021 at 4:14 PM John Keeping <john@...anate.com> wrote:

> GPIO nodes are not themselves busses, so passing rockchip_bank_match
> here is wrong.  Passing NULL instead uses the standard bus match table
> which is more appropriate.
>
> devm_of_platform_populate() shows that this is the normal way to call
> of_platform_populate() from a device driver, so in order to match that
> more closely also add the pinctrl device as the parent for the newly
> created GPIO controllers.
>
> Specifically, using the wrong match here can break dynamic GPIO hogs as
> marking the GPIO bank as a bus means that of_platform_notify() will set
> OF_POPULATED on new child nodes and if this happens before
> of_gpio_notify() is called then the new hog will be skipped as
> OF_POPULATED is already set.
>
> Fixes: 9ce9a02039de ("pinctrl/rockchip: drop the gpio related codes")
> Signed-off-by: John Keeping <john@...anate.com>

Patch applied.

Yours,
Linus Walleij