linux-kernel - Re: [PATCH V1] pinctrl: qcom: spmi-gpio: Add support for qcom,gpios-disallowed property

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:   Tue, 29 Aug 2017 09:51:13 +0800
From:   Shawn Guo <shawnguo@...nel.org>
To:     Fenglin Wu <fenglinw@...eaurora.org>
Cc:     linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org,
        bjorn.andersson@...aro.org,
        Linus Walleij <linus.walleij@...aro.org>,
        Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        linux-gpio@...r.kernel.org, devicetree@...r.kernel.org,
        collinsd@...eaurora.org, aghayal@...eaurora.org,
        wruan@...eaurora.org, kgunda@...eaurora.org
Subject: Re: [PATCH V1] pinctrl: qcom: spmi-gpio: Add support for
 qcom,gpios-disallowed property

On Tue, Aug 29, 2017 at 09:03:02AM +0800, Fenglin Wu wrote:
> I agree the GPIO's ownership is configurable and it always configured at
> the very beginning of the device boot up which is not visible by linux
> kernel drivers/image. Normally, this configuration is fixed in one
> platform and it's been protected and not allowed to be configured in
> linux kernel driver. So from linux driver point of view, this is a
> hardware configuration. I agree the coming patch "spmi: pmic-arb: Move
> the ownership check to irq_chip callback" would fix the pinctrl-
> spmi-gpio driver probe failure caused by the ownership mismatch, but
> this is just hiding the mistake of the kernel configured the GPIOs which
> not owned by APPS processor.

The kernel does everything just right, using the GPIO that device tree
tells to use.  If there is something wrong about ownership check, it
should be fault of that device tree specifies the wrong GPIO, or
firmware doesn't configure ownership as needed.

Shawn

> And these GPIOs will be registered
> successfully as pinctrl pins and any APPS processor consumer drivers
> could use this pins. This is not correct even the select_state operation
> for these pins would failed due to the mode protection in spmi write_cmd
> calling. I am thinking that not allowing these pins to be register as
> pinctrl pins should be more straightforward and easy understanding. So I
> think this patch still have value even the probe failure has been fixed
> by the coming spmi patch.