[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <917ac1f1-99a6-ac84-7486-ac97b0a617d8@ti.com>
Date: Mon, 16 Apr 2018 11:55:58 +0530
From: Sekhar Nori <nsekhar@...com>
To: David Lechner <david@...hnology.com>,
Linus Walleij <linus.walleij@...aro.org>
CC: Linux ARM <linux-arm-kernel@...ts.infradead.org>,
Kevin Hilman <khilman@...nel.org>,
Haojian Zhuang <haojian.zhuang@...aro.org>,
Tony Lindgren <tony@...mide.com>,
"open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS"
<devicetree@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ARM: dts: da850: use gpio-ranges
On Monday 16 April 2018 12:27 AM, David Lechner wrote:
> On 02/23/2018 05:44 AM, Sekhar Nori wrote:
>> On Thursday 22 February 2018 09:04 PM, Linus Walleij wrote:
>>> On Mon, Feb 19, 2018 at 11:29 PM, David Lechner
>>> <david@...hnology.com> wrote:
>>>
>>>> This makes use of the gpio-ranges feature that connects GPIO
>>>> controllers and PINMUX controllers.
>>>>
>>>> In da850.dtsi, pinctrl-single,gpio-range is added to the pinmux node
>>>> and gpio-ranges is added to the GPIO node. Unfortunately, the way the
>>>> pinctrl-single driver enumerates the pins (starting with LSB) causes
>>>> them to be in reverse order compared to the way the gpios are assigned.
>>>> As a result, we have to declare the mapping for each GPIO individually.
>>>>
>>>> This also lets us remove all of the GPIO pinmuxes from
>>>> da850-lego-ev3.dts. (Other da850 boards do not currently have any
>>>> GPIO pinmuxes declared.)
>>>>
>>>> Signed-off-by: David Lechner <david@...hnology.com>
>>>
>>> Nice,
>>> Acked-by: Linus Walleij <linus.wallei@...aro.org>
>>
>> Looks good to me too. And quite painstakingly done!
>>
>> I suppose this needs to wait a bit due to the dependencies? Let me know
>> once its okay to merge.
>>
>> Thanks,
>> Sekhar
>>
>
> Hi Sekhar,
>
> The dependencies are in v4.17-rc1 so you can pick up this patch now.
applied to v4.18/dt. I had to manually adjust the patch to make it
apply, probably because it was based off your common clock conversion
branch.
Thanks,
Sekhar
Powered by blists - more mailing lists