Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S933280AbbGJVLr (ORCPT <rfc822;w@1wt.eu>);
	Fri, 10 Jul 2015 17:11:47 -0400
Received: from devils.ext.ti.com ([198.47.26.153]:58096 "EHLO
	devils.ext.ti.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S932636AbbGJVLk (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 10 Jul 2015 17:11:40 -0400
Message-ID: <55A034F2.9010404@ti.com>
Date: Sat, 11 Jul 2015 06:11:14 +0900
From: "Kim, Milo" <milo.kim@ti.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.6.0
MIME-Version: 1.0
To: Sean Paul <seanpaul@chromium.org>
CC: Jingoo Han <jingoohan1@gmail.com>, Lee Jones <lee.jones@linaro.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2] backlight: lp855x: use private data for regulator
 control
References: <1436516798-4132-1-git-send-email-milo.kim@ti.com> <CAOw6vbJy0p7Hg+W4=3sddj1rTOXxf4v9S2Gp-fPnJq3rVD4Fdg@mail.gmail.com> <55A02E89.4090403@ti.com> <CAOw6vbL7epag+2GuR8yRsLxuio7v+OxwekS=33kVNTAUfBuyNQ@mail.gmail.com> <55A03086.7050207@ti.com>
In-Reply-To: <55A03086.7050207@ti.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 4143
Lines: 109

Hi Paul,

On 7/11/2015 5:52 AM, Kim, Milo wrote:
> Hi Paul,
>
> On 7/11/2015 5:49 AM, Sean Paul wrote:
>> On Fri, Jul 10, 2015 at 4:43 PM, Kim, Milo <milo.kim@ti.com> wrote:
>>> Hi Paul,
>>>
>>>
>>> On 7/11/2015 12:01 AM, Sean Paul wrote:
>>>>
>>>> On Fri, Jul 10, 2015 at 4:26 AM, Milo Kim <milo.kim@ti.com> wrote:
>>>>>
>>>>> LP855x backlight device can be enabled by external VDD input.
>>>>> The 'supply' data is used for this purpose.
>>>>> It's kind of private data which runs internally, so there is no reason to
>>>>> expose to the platform data.
>>>>>
>>>>> And devm_regulator_get() is moved from _parse_dt() to _probe().
>>>>> Regulator consumer(lp855x) can control regulator not only from DT but
>>>>> also
>>>>> from platform data configuration in a source file such like board-*.c.
>>>>>
>>>>> If 'power' regulator driver is not ready, lp855x should continue to work
>>>>> because the power supply control is optional. So -EPROBE_DEFER return
>>>>> code
>>>>> is removed.
>>>>>
>>>>> v1->v2:
>>>>>      Keeps optional property '<name>-supply' in LP855x DT binding.
>>>>>
>>>>> Cc: Sean Paul <seanpaul@chromium.org>
>>>>> Cc: Jingoo Han <jingoohan1@gmail.com>
>>>>> Cc: Lee Jones <lee.jones@linaro.org>
>>>>> Cc: linux-kernel@vger.kernel.org
>>>>> Signed-off-by: Milo Kim <milo.kim@ti.com>
>>>>> ---
>>>>>     drivers/video/backlight/lp855x_bl.c  | 20 +++++++++-----------
>>>>>     include/linux/platform_data/lp855x.h |  2 --
>>>>>     2 files changed, 9 insertions(+), 13 deletions(-)
>>>>>
>>>>> diff --git a/drivers/video/backlight/lp855x_bl.c
>>>>> b/drivers/video/backlight/lp855x_bl.c
>>>>> index a26d3bb..277d5ca 100644
>>>>> --- a/drivers/video/backlight/lp855x_bl.c
>>>>> +++ b/drivers/video/backlight/lp855x_bl.c
>>>>> @@ -73,6 +73,7 @@ struct lp855x {
>>>>>            struct device *dev;
>>>>>            struct lp855x_platform_data *pdata;
>>>>>            struct pwm_device *pwm;
>>>>> +       struct regulator *supply;       /* regulator for VDD input */
>>>>>     };
>>>>>
>>>>>     static int lp855x_write_byte(struct lp855x *lp, u8 reg, u8 data)
>>>>> @@ -384,13 +385,6 @@ static int lp855x_parse_dt(struct lp855x *lp)
>>>>>                    pdata->rom_data = &rom[0];
>>>>>            }
>>>>>
>>>>> -       pdata->supply = devm_regulator_get(dev, "power");
>>>>> -       if (IS_ERR(pdata->supply)) {
>>>>> -               if (PTR_ERR(pdata->supply) == -EPROBE_DEFER)
>>>>> -                       return -EPROBE_DEFER;
>>>>> -               pdata->supply = NULL;
>>>>> -       }
>>>>> -
>>>>>            lp->pdata = pdata;
>>>>>
>>>>>            return 0;
>>>>> @@ -431,8 +425,12 @@ static int lp855x_probe(struct i2c_client *cl, const
>>>>> struct i2c_device_id *id)
>>>>>            else
>>>>>                    lp->mode = REGISTER_BASED;
>>>>>
>>>>> -       if (lp->pdata->supply) {
>>>>> -               ret = regulator_enable(lp->pdata->supply);
>>>>> +       lp->supply = devm_regulator_get(lp->dev, "power");
>>>>> +       if (IS_ERR(lp->supply))
>>>>> +               lp->supply = NULL;
>>>>
>>>>
>>>>
>>>> Hi Milo,
>>>> You removed the probe deferral handling on the regulator and broke
>>>> probe deferral in cases where the regulator isn't ready.
>>>
>>>
>>> This power supply is optional. Even if lp855x can not get regulator driver,
>>> it should work. (And I saw same comment in the DT. The 'power-supply'
>>> property is optional). So -EPORBE_DEFER is not necessary in _probe().
>>>
>>
>> I respectfully disagree. devm_regulator_get can return EPROBE_DEFER if
>> the regulator is valid (and specified in the dt), but not ready to be
>> used yet. In this case, your patch will assume it doesn't exist and
>> will never use it. This Is Bad.
>
> So do you think this power supply should be mandatory in this driver?

The devm_regulator_get_optinonal() seems the right API for this case.
Let me take a look at and get back to you.

Best regards,
Milo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/