linux-kernel - Re: [PATCH v6 04/10] regulator: fixed: add support for ACPI interface

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, 27 Apr 2016 09:54:10 +0800
From:	Lu Baolu <baolu.lu@...ux.intel.com>
To:	Mark Brown <broonie@...nel.org>
Cc:	felipe.balbi@...ux.intel.com,
	Mathias Nyman <mathias.nyman@...el.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Lee Jones <lee.jones@...aro.org>,
	Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
	MyungJoo Ham <myungjoo.ham@...sung.com>,
	Chanwoo Choi <cw00.choi@...sung.com>,
	Liam Girdwood <lgirdwood@...il.com>, linux-usb@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 04/10] regulator: fixed: add support for ACPI interface

Hi,

On 04/26/2016 06:23 PM, Mark Brown wrote:
> On Tue, Apr 26, 2016 at 10:24:56AM +0800, Lu Baolu wrote:
>
>> The GPIO name might be different in different use cases. For my case,
>> it is "vbus_en", but other cases should use the different name.
>> On ACPI compatible platforms, GPIO resources are reported via ACPI
>> tables and (devm_)gpiod_get() hides the APCI complexity and returns
>> the gpiod according to "gpio_name".
> That's labelling that you might want to do on the supplier side or at
> system level.

The labeling is done at firmware level (ACPI 5.1). It uses _DSD
configuration object to give names to GPIOs. There are systems
which don't contain _DSD. On those platforms, Linux kernel
could do this instead.

Please refer to Documentation/acpi/gpio-properties.txt.

> Why does the device care?It's requesting the GPIO in
> its own context and it's only requesting one GPIO, with DT we're just
> always calling the GPIO "gpio" which works fine.

This driver is not bound to an ACPI device node directly. It's a child
of a mfd device, which is corresponding to a real ACPI device node.

I agree with you that we should not retrieve gpio name from the
device provider. Driver should have the knowledge of the gpio name.
(Please correct me if I didn't understand your point right. :-) )

Best regards,
Lu Baolu