[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170718091906.trvcaacdfrizjfks@dell>
Date: Tue, 18 Jul 2017 10:19:06 +0100
From: Lee Jones <lee.jones@...aro.org>
To: Enric Balletbo Serra <eballetbo@...il.com>
Cc: Gwendal Grignou <gwendal@...omium.org>,
Enric Balletbo i Serra <enric.balletbo@...labora.com>,
Jonathan Cameron <jic23@...nel.org>,
Benson Leung <bleung@...omium.org>,
Javier Martinez Canillas <martinez.javier@...il.com>,
Guenter Roeck <groeck@...omium.org>,
Linux Kernel <linux-kernel@...r.kernel.org>,
linux-iio@...r.kernel.org, rtc-linux@...glegroups.com
Subject: Re: [PATCH v3 1/4] mfd: cros_ec: Get rid of cros_ec_check_features
from cros_ec_dev.
On Mon, 17 Jul 2017, Enric Balletbo Serra wrote:
> Hi Gwendal,
>
> 2017-07-13 22:33 GMT+02:00 Gwendal Grignou <gwendal@...omium.org>:
> > On Wed, Jul 12, 2017 at 3:13 AM, Enric Balletbo i Serra
> > <enric.balletbo@...labora.com> wrote:
> >> The cros_ec_dev driver should be used only to expose the Chrome OS Embedded
> >> Controller to user-space and should not be used to add MFD devices by
> >> calling mfd_add_devices. This patch moves this logic to the MFD cros_ec
> >> driver and removes the MFD bits from the character device driver. Also
> >> makes independent the IIO driver from the character device as also has no
> >> sense.
> >
> > cros_ec_dev serves another purpose: it allows to represent an EC that
> > does not have a cros_ec structure. It happens when there are several
> > EC in a chromebook, and one EC is connected through another EC.
> > One example is Samus (Pixel 2): where we have:
> >
> > (main SOC, Application Processor) AP --> (main Embedded Controller) EC
> > ---> (Power Delivery [PD}) EC
> >
> > We access to the PD EC via pass-through commands through the main EC.
> > Each EC has a cros_ec_dev structure, but only the main EC as a
> > cros_ec_device structure (I will forever regret the structure names).
> >
> > Now form the AP point of view, both ECs use the same protocol. That
> > why the sensors and other devcies that are registered by looking at
> > the feature fields are registered with cros_ec_dev as their parent.
> > Other devices that are registered from the device tree, predating the
> > feature field support, are registered with cros_ec_device as their
> > parent.
> >
>
> Interesting I didn't know that. So are you saying that this patch will
> break support for devices like Pixel 2? I tested the patches on
> various devices but not on Pixel 2 so could be.
Does this affect the rest of the series?
Or should I review/apply as usual?
--
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
Powered by blists - more mailing lists