Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH V7 2/2] iio: accell: mma8452: add vdd/vddio regulator
 operation support
To:     Jonathan Cameron <jic23@kernel.org>,
        Anson Huang <anson.huang@nxp.com>
Cc:     "knaack.h@gmx.de" <knaack.h@gmx.de>,
        "lars@metafoo.de" <lars@metafoo.de>,
        "pmeerw@pmeerw.net" <pmeerw@pmeerw.net>,
        "robh+dt@kernel.org" <robh+dt@kernel.org>,
        "mark.rutland@arm.com" <mark.rutland@arm.com>,
        "gustavo@embeddedor.com" <gustavo@embeddedor.com>,
        "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
        Leonard Crestez <leonard.crestez@nxp.com>,
        "rtresidd@electromag.com.au" <rtresidd@electromag.com.au>,
        "linux-iio@vger.kernel.org" <linux-iio@vger.kernel.org>,
        "devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        dl-linux-imx <linux-imx@nxp.com>
References: <1546938556-26969-1-git-send-email-Anson.Huang@nxp.com>
 <1546938556-26969-2-git-send-email-Anson.Huang@nxp.com>
 <84c2971b-29d5-a0d0-c830-11504c381dd2@posteo.de>
 <4CB70D1B-06A6-4E34-B701-58E72B1AC9D9@nxp.com>
 <20190112191111.4909e254@archlinux>
From:   Martin Kepplinger <martink@posteo.de>
Openpgp: preference=signencrypt
Autocrypt: addr=martink@posteo.de; keydata=
 mQINBFULfZABEADRxJqDOYAHfrp1w8Egcv88qoru37k1x0Ugy8S6qYtKLAAt7boZW+q5gPv3
 Sj2KjfkWA7gotXpASN21OIfE/puKGwhDLAySY1DGNMQ0gIVakUO0ji5GJPjeB9JlmN5hbA87
 Si9k3yKQQfv7Cf9Lr1iZaV4A4yjLP/JQMImaCVdC5KyqJ98Luwci1GbsLIGX3EEjfg1+MceO
 dnJTKZpBAKd1J7S2Ib3dRwvALdiD7zqMGqkw5xrtwasatS7pc6o/BFgA9GxbeIzKmvW/hc3Q
 amS/sB12BojyzdUJ3TnIoAqvwKTGcv5VYo2Z+3FV+/MJVXPo8cj2vmfxQx1WG4n6X0pK4X8A
 BkCKw2N/evMZblNqAzzGVtoJvqQYkzQ20Fm+d3wFl6lS1db4MB+kU13G8kEIE22Q3i6kx4NA
 N49FLlPeDabGfJUyDaZp5pmKdcd7/FIGH/HjShjx7g+LKSwWNMkDygr4WARAP4h8zYDZuNqe
 ofPvMLqJxHeexBPIGF/+OwMyTvM7otP5ODuFmq6OqjNPf1irJmkiFv3yEa+Ip0vZzwl4XvrZ
 U0IKjSy2rbRLg22NsJT0XVZJbutIXYSvIHGqSxzzfiOOLnRjR++fbeEoVlRJ4NZHDKCh3pJv
 LNd+j03jXr4Rm058YLgO7164yr7FhMZniBJw6z648rk8/8gGPQARAQABtCVNYXJ0aW4gS2Vw
 cGxpbmdlciA8bWFydGlua0Bwb3N0ZW8uZGU+iQI6BBMBAgAkAhsDAh4BAheABQsJCAcDBRUK
 CQgLBRYCAwEABQJVC4DBAhkBAAoJEFADmN9as4fTpYwQAIqwZ2arvCsfwiZqr/KyJ4ewhn2/
 7JVR/kvx5G6nfPI55XtNDmd2Lt7xNvY5LbLwGp2c3JMD1rZ2FhbWXC39SA0yxeE4U0NTlxDg
 RGx20k85pZTFvxyPfz9c7dAFTLMajpzLvpjBjEaqVm6KnS/UBBaGHOu0999siD1EDaSBWUiO
 HPMXNYkcFt96p55LYNAgzSsd+zTjknxCnmzUMiDKzjFn6LdqdlyPyMj6IXpeiAFHV43SAGb6
 8miE+S61pq9pTapt+E5qf3zfuKATK0dfZkkMFaC+Vmv6DvcpR7G1ilpmjkR6o/mDM6dtm21T
 5jpYrEmb7hgigFl9Pg01mJLwSGm1GYf45aKQH/VZff+sYsDDNQUHwabG9DVV/edSRJGzCu3R
 W/xqeF3Ll44Bhaa9LaVQuN7Yuqixhxm8flJNcfnknYd9TBQYLIZLcUyN3bbaABbCv6xkHaB6
 ZUUQPhpVGoLANrLtTSEtYBYzktSmeARLTtVt5wJ0Q8gQ6h5a0VC6zHv37cRUYqsEwwRwbG+h
 aBs907W8hH4etQtbbXBbbbXnOOl/QnpShjyWYe02A/f/QWpgZD5SPsB6RVQdWnP8ZN7OngzE
 RACA2ftyBnp/0ESKMDLYJDRGm3oM01hZSZHnFBt/aggx3FOM39bmu565xg21hO7I7s9xkvbZ
 Czz2iSRTuQINBFULfZABEADFNrM9n2N+nq4L4FKIi2PCSsWWU0RUqm26b3wkmi9anWSJsz6m
 GXqJWj7AoV6w2ybnry+IzYIDN7NWUyvsXS7o1A0rqm7Tzhb3IdJQpE4UWvzdSKfq3ThTzy1w
 KIFgtDkb5OtW4Zf/mpjV6tVYjjJx2SpDNvwA9swWtb+xFvvzV/zAZdaEOzoF3g81goe/sLSv
 xdijvs95KoZJX/nmWlKyagTb7NHcxblNWhoTzdnGF+qC1MhYx/zyaD/bQQiFgJEbSI6aNfK1
 Z/77Eub3Gkx4qcp9ZdDFFt+8qDf4rMXfQDSE7dgHIoQ1ifC1IHPyh3fY3uicbn75rPF+6Fhk
 bkyRo14k8so9CnIYxzY+ienQGEJlO/EhsjzVl5fpML45lt5b7TeIacLsSjjIn3dBSTNYU6EY
 YTHQUeP6oGQNAuxEQRjCx3Gqqv2TUpQPUYVUOXSDO4qqJXhiOUmIV8eH19tMPO2vc2X+tpY0
 3EDcy1f2ey06vtv4+gDiAfUZcv1hKVd18E9WeuGCm64lhyovLTaLf/3RSSKL33SeaLkLPOEF
 UXA2OxlNfDs1FK0is+0oJr55ZEI7N9o6oFQp+bNcQeAyXh6yqTIW7YxK9tHpyUhVqOQGZzj5
 0SC/XdEn1VZbqo11DDupNsMlp+BBRuY5QwjKANGMIAvay38uICLYxaCXzQARAQABiQIfBBgB
 AgAJBQJVC32QAhsMAAoJEFADmN9as4fTBJkQAKl9A9gUvgiLgilK6OoR9vX+cv4yL7c0uubw
 eneL+ZWAytTAF3jHT6cPFzv4rD8iJc1yhAFDc0LW+yywnoP7Tok6cYlYH1DCjIQsZ1Du1Jad
 rjTmvAPFyzKc2dcNPR3f1DAU3adcLLKz7v4+uLmBPI4HIn4TnYXbttfb0vTmJVJFERV7XMsu
 NiQVDgsM1K1Sn9xqYPoU59v725VzOwyhNnV2jZC2MkyVGWFKEbPcZhTDnaFpYp83e2y+sgeN
 l/YXkBjLnM4SCt/w7eObYsM2J2KfzfT5QdtqglWJsJMm91tWqn8GUDUgqnWz9jzzKVKDEMXA
 W5dQSUkD0aWY0cDNkFqs8QlWRgFMelG0gqnCqZRMf/IfSnN23yGK0j5EENjKdifSdTGItlQ8
 B4znBEu3VdpDZANzRAlHxXAEJVJ7z7fmAQ9079CauV43mIDeo4cxbxfBcmiR3sxpLoUkoZ0W
 ONk8MxHhCLw9OfYubU2QMekS1oSOMqZ2u3/g6kTp9XiIq0LWRy862+rE1fOYWf3JpsdWVszB
 NjZPEXwiZ9m+v/VJ3NuzrLOJqw1F/FMaaZgbauYH9c7oAx1qXl7BYMV9WYiJGiJV0xK5UzpD
 GsOfIJ8/tbwPSs6pNZDAJata///+/Py99NtaU3bUYhyluAGZ/2UHygGkuyZnJc2mWFBWYWWi
 uQINBFu90XkBEADNiGnoPNJM51vHkoR5hOFtGs0TC3WDMJughEcqfrsNkExX5pL3ce9mNXvH
 XE283sLO0ZAhzhMShg/z7jrZyX8zNeYJdUhteSqHH5ThHky/xGpolv6cE+O8hKDCjmegzzlG
 PhjizCCZilCmGh2hqQqgcpAw7j7odEf+PORSqgyDZ6IDdm7rbHLjy0Gu+5PlEuUoyRYpS3Uu
 w+DKZUXaVDa6wVYYQnulMBTyucEFU/SJojOgxv8UfWjHvcOumA+2xsl8AXgB4KOiXhuGPeMZ
 A1Bg5iFMyZhXTxwyPVIl81xgunO7N5qpxh+c6A2tm+i+8/nqbhdJwi+8qBOMemb8KHLnCfjk
 5WKn92aeDbCbD+0wfHSsUqYE/F47VFC6sEddVvuXiYEp6HBjgtvfAEUeJcIaxqVy0/xonRx9
 HsIlljNvadrk0T+DX4Uz65z3buOSdU6eOAbGpG4MFGEcuef2uRBzd3AlbVg9DH4fpd3kjVSR
 5qere3Aer1r+40kFrhOZpBtI9zg5Oea7p4MumFNPRG+9MCyAgLA1iGoaZcN6gLcrGTYZ98SA
 BDJJWoc7PkHsxJ0lGAnWg12fYihRE5NHYLewdlacNkhKUHMeEiv9zXUZdnRBESKznjZf2+z6
 xcrc/IUPH9KNL/53Nxf41qEw3id5sKA4ef1+SapHaiJm0SP0vQARAQABiQRyBBgBCAAmFiEE
 8ggriA+eQjk0aG4/UAOY31qzh9MFAlu90XkCGwIFCQHhM4ACQAkQUAOY31qzh9PBdCAEGQEI
 AB0WIQTMkt66NxM4ndQ+Z3eJLhmpqZPCyQUCW73ReQAKCRCJLhmpqZPCyXufEADGZ1T15vbC
 AVxUf54H1Hm2PRg3zpypa9abDSeXtKp9eyADi3O7WB/hwkTapP/lilmCs/RBZ6hJm+re+5Qh
 wAId5aVYd5XYrFRR5IMu8B4hanXfos2gzjDvfPLRSscUo+TjELgy0jR6Bz4/RXopyeDuYOHv
 p6WQ0ZyyXVEqRmfigLMRAy2tZKb9d1ovjxQZEkDLpmJTN7yp32tX4cI0CkcUzkHmRhvzQ3Vd
 tMxo2QPzKHJ7k0w2xbGpmChT2+oTf03g03eJ2SLmqYTIMrquMtPejPukPBOpnNM0K8az3shw
 uMYLLWPYbd8V2HkopQ+8XyV9WPrquVSWoYxPRztPxsNeaVGRlfx9Cdy38Usf1dhUJ3OWu1LJ
 YeIAbxkVX3z9ME77Jg9dHxdDT79r0X9AuPJJ0vKMUrGrIsiWzAXTJudR6VOXSXbc+HN+om96
 zjQUXf6OwD+N8e49L8klDK+OypyXQwt0rjiaxF+otgTZchCBDMqfE/0IIcQomYCpAqNrnkCm
 cVsmirN5hXedSN2tzMmxePBFsTfrq/sdP/hJeTMUcwn/lnC5DgygnXU6SazoS+e8lTGb4JJ5
 Q74XKS7LODpu3eC5uifnCxq6M5CYQOWgD9aGdWEGEViIcJndzC+h/+2plSvGyRJgJoj+pbiC
 UgYSujU0HH8gXVAbD+1pNgfXSg4hD/4qNFv91G8Et2DThur4wZJxAfAmHykmMxxQnJ/2+VA1
 Ej+QrsU3sq93D8gJNEVVV96tcGrJv4ytyRLmt5fi35xEvdxgf7FiS3bQqp9rS6jL/lhz7qNw
 mR8c4pgMAdURioI8QnPlgSEfneH8KVCWoCc09kysjG1rmVh2dXpIQo3Shri/zhbWtVQepTLI
 D+VjnhdFPVXTixI1b4NW6h4aSkrip6ccyH/8WjsGqPQVX5yNSAJnoqjlpb3buVEU+EP9uvg/
 RvNj6qFRbc+nfOBBgrUvhEr0aEOvFotExGvNjR5T9MewbGVnL4s0IlMYwZBD88ygm77CmQRg
 MbBzOP6HyiLOUmLBbfDXmHP0ChP3Oi34bm7OSPq9PyxcuMhr1hzWXQYsjqH6CtU0cNRjeIz/
 bEt7TFPkP0uzJ4zLjX2BAJ9WJC2JGDWEMhtzg2kZaAS9jwCEAf7m8mP+RT64rSMztjKtmIQN
 vZicCb2MBQaC7vS7thkGhyQIlzN32iWeCHeydZnDPMEU9PP3PMsQHDAnZdXjc5Jfa2q9WOzG
 nCFhEis48yoFOobYJHR0viwuY4DIprx3Dq6I4+WVwuf6U6y579i+qDvOxiuFcClGgBmsmehF
 YSKzZPSgZYhCVZhzSpg4D+AVQEhxhGTSCXDSBIMpnW4GgdJdZGmBBobKqvHmgI95tw==
Message-ID: <3540a0d9-b756-af00-65b8-c78ba669ebff@posteo.de>
Date:   Sun, 13 Jan 2019 09:31:22 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.4.0
MIME-Version: 1.0
In-Reply-To: <20190112191111.4909e254@archlinux>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 12.01.19 20:11, Jonathan Cameron wrote:
> On Tue, 8 Jan 2019 14:48:48 +0000
> Anson Huang <anson.huang@nxp.com> wrote:
> 
>> Hi, Martin
>>
>> From Anson's iPhone 6
>>
>>
>>> 在 2019年1月8日，19:41，Martin Kepplinger <martink@posteo.de> 写道：
>>>   
>>>> On 08.01.19 10:14, Anson Huang wrote:
>>>> The accelerometer's power supply could be controllable on some
>>>> platforms, such as i.MX6Q-SABRESD board, the mma8451's power supplies
>>>> are controlled by a GPIO fixed regulator, need to make sure the
>>>> regulators are enabled before any communication with mma8451, this
>>>> patch adds vdd/vddio regulator operation support.
>>>>
>>>> Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
>>>> Acked-by: Martin Kepplinger <martink@posteo.de>
>>>> ---
>>>> ChangeLog Since V6:
>>>>  - separate the error handling of regulators get to make code easy to read.
>>>> ---
>>>> drivers/iio/accel/mma8452.c | 105 ++++++++++++++++++++++++++++++++++----------
>>>> 1 file changed, 83 insertions(+), 22 deletions(-)
>>>>
>>>> diff --git a/drivers/iio/accel/mma8452.c b/drivers/iio/accel/mma8452.c
>>>> index 421a0a8..3027811 100644
>>>> --- a/drivers/iio/accel/mma8452.c
>>>> +++ b/drivers/iio/accel/mma8452.c
>>>> @@ -31,6 +31,7 @@
>>>> #include <linux/of_device.h>
>>>> #include <linux/of_irq.h>
>>>> #include <linux/pm_runtime.h>
>>>> +#include <linux/regulator/consumer.h>
>>>>
>>>> #define MMA8452_STATUS                0x00
>>>> #define  MMA8452_STATUS_DRDY            (BIT(2) | BIT(1) | BIT(0))
>>>> @@ -107,6 +108,8 @@ struct mma8452_data {
>>>>    u8 data_cfg;
>>>>    const struct mma_chip_info *chip_info;
>>>>    int sleep_val;
>>>> +    struct regulator *vdd_reg;
>>>> +    struct regulator *vddio_reg;
>>>> };
>>>>
>>>>  /**
>>>> @@ -1534,9 +1537,39 @@ static int mma8452_probe(struct i2c_client *client,
>>>>    mutex_init(&data->lock);
>>>>    data->chip_info = match->data;
>>>>
>>>> +    data->vdd_reg = devm_regulator_get(&client->dev, "vdd");
>>>> +    if (IS_ERR(data->vdd_reg)) {
>>>> +        if (PTR_ERR(data->vdd_reg) == -EPROBE_DEFER)
>>>> +            return -EPROBE_DEFER;
>>>> +
>>>> +        dev_err(&client->dev, "failed to get VDD regulator!\n");
>>>> +        return PTR_ERR(data->vdd_reg);
>>>> +    }
>>>> +
>>>> +    data->vddio_reg = devm_regulator_get(&client->dev, "vddio");
>>>> +    if (IS_ERR(data->vddio_reg)) {
>>>> +        if (PTR_ERR(data->vddio_reg) == -EPROBE_DEFER)
>>>> +            return -EPROBE_DEFER;
>>>> +
>>>> +        dev_err(&client->dev, "failed to get VDDIO regulator!\n");
>>>> +        return PTR_ERR(data->vddio_reg);
>>>> +    }
>>>> +
>>>> +    ret = regulator_enable(data->vdd_reg);
>>>> +    if (ret) {
>>>> +        dev_err(&client->dev, "failed to enable VDD regulator!\n");
>>>> +        return ret;
>>>> +    }
>>>> +
>>>> +    ret = regulator_enable(data->vddio_reg);
>>>> +    if (ret) {
>>>> +        dev_err(&client->dev, "failed to enable VDDIO regulator!\n");
>>>> +        goto disable_regulator_vdd;
>>>> +    }
>>>> +
>>>>    ret = i2c_smbus_read_byte_data(client, MMA8452_WHO_AM_I);
>>>>    if (ret < 0)
>>>> -        return ret;
>>>> +        goto disable_regulators;
>>>>
>>>>    switch (ret) {
>>>>    case MMA8451_DEVICE_ID:
>>>> @@ -1549,7 +1582,8 @@ static int mma8452_probe(struct i2c_client *client,
>>>>            break;
>>>>        /* else: fall through */
>>>>    default:
>>>> -        return -ENODEV;
>>>> +        ret = -ENODEV;
>>>> +        goto disable_regulators;
>>>>    }
>>>>
>>>>    dev_info(&client->dev, "registering %s accelerometer; ID 0x%x\n",
>>>> @@ -1566,13 +1600,13 @@ static int mma8452_probe(struct i2c_client *client,
>>>>
>>>>    ret = mma8452_reset(client);
>>>>    if (ret < 0)
>>>> -        return ret;
>>>> +        goto disable_regulators;
>>>>
>>>>    data->data_cfg = MMA8452_DATA_CFG_FS_2G;
>>>>    ret = i2c_smbus_write_byte_data(client, MMA8452_DATA_CFG,
>>>>                    data->data_cfg);
>>>>    if (ret < 0)
>>>> -        return ret;
>>>> +        goto disable_regulators;
>>>>
>>>>    /*
>>>>     * By default set transient threshold to max to avoid events if
>>>> @@ -1581,7 +1615,7 @@ static int mma8452_probe(struct i2c_client *client,
>>>>    ret = i2c_smbus_write_byte_data(client, MMA8452_TRANSIENT_THS,
>>>>                    MMA8452_TRANSIENT_THS_MASK);
>>>>    if (ret < 0)
>>>> -        return ret;
>>>> +        goto disable_regulators;
>>>>
>>>>    if (client->irq) {
>>>>        int irq2;
>>>> @@ -1595,7 +1629,7 @@ static int mma8452_probe(struct i2c_client *client,
>>>>                        MMA8452_CTRL_REG5,
>>>>                        data->chip_info->all_events);
>>>>            if (ret < 0)
>>>> -                return ret;
>>>> +                goto disable_regulators;
>>>>
>>>>            dev_dbg(&client->dev, "using interrupt line INT1\n");
>>>>        }
>>>> @@ -1604,11 +1638,11 @@ static int mma8452_probe(struct i2c_client *client,
>>>>                    MMA8452_CTRL_REG4,
>>>>                    data->chip_info->enabled_events);
>>>>        if (ret < 0)
>>>> -            return ret;
>>>> +            goto disable_regulators;
>>>>
>>>>        ret = mma8452_trigger_setup(indio_dev);
>>>>        if (ret < 0)
>>>> -            return ret;
>>>> +            goto disable_regulators;
>>>>    }
>>>>
>>>>    data->ctrl_reg1 = MMA8452_CTRL_ACTIVE |
>>>> @@ -1661,12 +1695,19 @@ static int mma8452_probe(struct i2c_client *client,
>>>> trigger_cleanup:
>>>>    mma8452_trigger_cleanup(indio_dev);
>>>>
>>>> +disable_regulators:
>>>> +    regulator_disable(data->vddio_reg);
>>>> +
>>>> +disable_regulator_vdd:
>>>> +    regulator_disable(data->vdd_reg);
>>>> +
>>>>    return ret;
>>>> }
>>>>
>>>> static int mma8452_remove(struct i2c_client *client)
>>>> {
>>>>    struct iio_dev *indio_dev = i2c_get_clientdata(client);
>>>> +    struct mma8452_data *data = iio_priv(indio_dev);
>>>>
>>>>    iio_device_unregister(indio_dev);
>>>>
>>>> @@ -1678,6 +1719,9 @@ static int mma8452_remove(struct i2c_client *client)
>>>>    mma8452_trigger_cleanup(indio_dev);
>>>>    mma8452_standby(iio_priv(indio_dev));
>>>>
>>>> +    regulator_disable(data->vddio_reg);
>>>> +    regulator_disable(data->vdd_reg);
>>>> +
>>>>    return 0;
>>>> }
>>>>
>>>> @@ -1696,6 +1740,18 @@ static int mma8452_runtime_suspend(struct device *dev)
>>>>        return -EAGAIN;
>>>>    }
>>>>
>>>> +    ret = regulator_disable(data->vddio_reg);
>>>> +    if (ret) {
>>>> +        dev_err(dev, "failed to disable VDDIO regulator\n");
>>>> +        return ret;
>>>> +    }
>>>> +
>>>> +    ret = regulator_disable(data->vdd_reg);
>>>> +    if (ret) {
>>>> +        dev_err(dev, "failed to disable VDD regulator\n");
>>>> +        return ret;
>>>> +    }
>>>> +
>>>>    return 0;
>>>> }
>>>>
>>>> @@ -1705,9 +1761,22 @@ static int mma8452_runtime_resume(struct device *dev)
>>>>    struct mma8452_data *data = iio_priv(indio_dev);
>>>>    int ret, sleep_val;
>>>>
>>>> +    ret = regulator_enable(data->vdd_reg);
>>>> +    if (ret) {
>>>> +        dev_err(dev, "failed to enable VDD regulator\n");
>>>> +        return ret;
>>>> +    }
>>>> +
>>>> +    ret = regulator_enable(data->vddio_reg);
>>>> +    if (ret) {
>>>> +        dev_err(dev, "failed to enable VDDIO regulator\n");
>>>> +        regulator_disable(data->vdd_reg);
>>>> +        return ret;
>>>> +    }
>>>> +
>>>>    ret = mma8452_active(data);
>>>>    if (ret < 0)
>>>> -        return ret;
>>>> +        goto runtime_resume_failed;
>>>>
>>>>    ret = mma8452_get_odr_index(data);
>>>>    sleep_val = 1000 / mma8452_samp_freq[ret][0];
>>>> @@ -1717,25 +1786,17 @@ static int mma8452_runtime_resume(struct device *dev)
>>>>        msleep_interruptible(sleep_val);
>>>>
>>>>    return 0;
>>>> -}
>>>> -#endif
>>>>
>>>> -#ifdef CONFIG_PM_SLEEP
>>>> -static int mma8452_suspend(struct device *dev)
>>>> -{
>>>> -    return mma8452_standby(iio_priv(i2c_get_clientdata(
>>>> -        to_i2c_client(dev))));
>>>> -}
>>>> +runtime_resume_failed:
>>>> +    regulator_disable(data->vddio_reg);
>>>> +    regulator_disable(data->vdd_reg);
>>>>
>>>> -static int mma8452_resume(struct device *dev)
>>>> -{
>>>> -    return mma8452_active(iio_priv(i2c_get_clientdata(
>>>> -        to_i2c_client(dev))));
>>>> +    return ret;
>>>> }
>>>> #endif
>>>>
>>>> static const struct dev_pm_ops mma8452_pm_ops = {
>>>> -    SET_SYSTEM_SLEEP_PM_OPS(mma8452_suspend, mma8452_resume)
>>>> +    SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume)  
>>>
>>> Just a quesion: Isn't it semantically a different change to replace
>>> mma8452_suspend() and mma8452_resume()? If not, why is it needed here?
>>>
>>> other than that, lgtm.  
>>
>> It is because with regulator operation added, before system suspend, driver is in runtime suspend already, the suspend/resume callback need to deal with regulator again including error handling etc., and also need to disable runtime pm and re-enable it etc., this introduces too many complicated code/logic, also, the things suspend/resume callback does are actually same as runtime suspend/ resume, so just using pm_runtime_force_suspend/resume can save these duplicated operations and make code easy/clean.
>>
> Anson, please wrap below 80 chars.  Some of us have old old old
> laptops, (though I admit I still have more than 80 chars ;)
> 
> I 'think' Anson is correct that these end up running the same
> code, but with considerably less complexity an this is
> the reason the force functions exist.  Often the runtime case
> is very similar to the full pm case so you can avoid the whole
> runtime resume then suspend dance by doing this.
> 
> Anyhow I've applied it to the togreg branch of iio.git and pushed
> it out as testing.  Hopefully we haven't missed any corner
> cases.
> 
> Runtime PM always gives me a headache when I dig into exactly
> what the various dances are that go on.
> 

Alright. If not too late:

Acked-by: Martin Kepplinger <martink@posteo.de>

thanks,

                                 martin