Re: [PATCH v5 0/7] platform/chrome: Add user-space dev inferface support

From: Olof Johansson <olof@lixom.net>
To: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Cc: Lee Jones <lee.jones@linaro.org>,
	Doug Anderson <dianders@chromium.org>,
	Bill Richardson <wfrichar@chromium.org>,
	Simon Glass <sjg@google.com>,
	Gwendal Grignou <gwendal@google.com>,
	Jonathan Corbet <corbet@lwn.net>,
	Varka Bhadram <varkabhadram@gmail.com>,
	Paul Bolle <pebolle@tiscali.nl>,
	linux-samsung-soc@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v5 0/7] platform/chrome: Add user-space dev inferface support
Date: Wed, 25 Feb 2015 16:57:41 -0800	[thread overview]
Message-ID: <20150226005741.GB3101@quad.lixom.net> (raw)
In-Reply-To: <54E1A825.7090408@collabora.co.uk>

On Mon, Feb 16, 2015 at 09:19:49AM +0100, Javier Martinez Canillas wrote:
> Hello Olof,
> 
> On 02/02/2015 12:26 PM, Javier Martinez Canillas wrote:
> > Hello,
> > 
> > The mainline ChromeOS Embedded Controller (EC) driver is still missing some
> > features that are present in the downstream ChromiumOS tree. These are:
> > 
> >   - Low Pin Count (LPC) interface
> >   - User-space device interface
> >   - Access to vboot context stored on a block device
> >   - Access to vboot context stored on EC's nvram
> >   - Power Delivery Device
> >   - Support for multiple EC in a system
> > 
> > This is a fifth version of a series that adds support for the first two of
> > the missing features: the EC LPC and EC character device interfaces that
> > are used by user-space to access the ChromeOS EC. The support patches were
> > taken from the downstream ChromiumOS 3.14 tree with the fixes and cleanups
> > squashed to have a minimal patch-set.
> > 
> 
> Any comments on this series? The last version was posted a couple of weeks
> ago but the series have been in the list for months. Lee has already acked
> the mfd changes so you can merge all through your chrome-platform tree if
> you want.
> 
> It wold be great if this series get in to have the EC user-space interface
> supported and to minimize the delta with the Chromemium OS kernel since it
> still has other features that needs to be upstreamed like multiple EC in a
> system and access to vboot context stored in block device or EC's nvram.

Yeah, sorry for dragging my feet on this. The only thing I found that should
should be revisited is the reuse of the ioctl numbers to make it easier to
transition on the Chrome OS side -- otherwise it'll be hard to know from
userspace to use old or new structs during transition without a flag day.

-Olof