[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <7a18520a-34c8-2225-d4cd-b6d2522a2fe5@mellanox.com>
Date: Wed, 17 May 2017 18:07:24 -0400
From: David Woods <dwoods@...lanox.com>
To: Andy Shevchenko <andy.shevchenko@...il.com>,
Ulf Hansson <ulf.hansson@...aro.org>
CC: Jaehoon Chung <jh80.chung@...sung.com>,
Shawn Lin <shawn.lin@...k-chips.com>,
Heiner Kallweit <hkallweit1@...il.com>,
Adrian Hunter <adrian.hunter@...el.com>,
<cmetcalf@...lanox.com>,
"linux-mmc@...r.kernel.org" <linux-mmc@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] Use device_property_read instead of of_property_read in
the dw_mmc driver
On 05/15/2017 09:30 AM, Andy Shevchenko wrote:
> On Mon, May 15, 2017 at 12:36 PM, Ulf Hansson <ulf.hansson@...aro.org> wrote:
>> On 25 April 2017 at 17:39, David Woods <dwoods@...lanox.com> wrote:
>>> Using the device_property interfaces allows the dw_mmc driver to work
>>> on platforms which run on either device tree or ACPI.
>> Can you please split this to two separate changes. One for dw_mmc with
>> a proper prefix (mmc: dw_mmc:) and one for the mmc core (prefix "mmc:
>> core:").
Yes, I will do that.
> Btw, for the core part of change I would recommend to check other API
> functions which might go slightly different in case of ACPI vs DT vs
> built-in device properties.
>
> For example, if any code is using of_gpio header and of_ related stuff
> it should be rethought as well (not sure if we would change there
> anything, though worth to double check).
>
I have not purged all references to of_ APIs from the core code.
However, the ones which remain are specific to hardware I don't have, so
I wouldn't be able to test those code paths. Btw, there is no good
reason I can see to include of_gpio.h here, none of those interfaces
seem to be used.
-David Woods
Powered by blists - more mailing lists