From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.6 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,INCLUDES_PATCH,MAILING_LIST_MULTI, MENTIONS_GIT_HOSTING,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED, USER_AGENT_SANE_2 autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9D2D0C74A44 for ; Sun, 14 Jul 2019 17:14:37 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 6499D2089C for ; Sun, 14 Jul 2019 17:14:37 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1563124477; bh=M9BGjFEZHbJ2DxMD0K/QLdRqjF6rs+RkbEl1GoULyVs=; h=Date:From:To:Cc:Subject:In-Reply-To:References:List-ID:From; b=Y1V+bXHJ9Pg+Fv7/Hy5GM13npYlr/NHx4rIKQyIIJQU7p38X2QRFHSXCn496702cN YltU3JlilvxH8J8RhfI4ndQVA+JuBiQAkLr2p+0zkni89yq+KRd7WJltbxRZNw3H5j ZYR2xpAI6mnocLXLDZL823qkEqO1DVMtIzoEbuk0= Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728689AbfGNROg (ORCPT ); Sun, 14 Jul 2019 13:14:36 -0400 Received: from mail.kernel.org ([198.145.29.99]:39786 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728164AbfGNROf (ORCPT ); Sun, 14 Jul 2019 13:14:35 -0400 Received: from archlinux (cpc149474-cmbg20-2-0-cust94.5-4.cable.virginm.net [82.4.196.95]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id E0EA1205F4; Sun, 14 Jul 2019 17:14:31 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1563124473; bh=M9BGjFEZHbJ2DxMD0K/QLdRqjF6rs+RkbEl1GoULyVs=; h=Date:From:To:Cc:Subject:In-Reply-To:References:From; b=Qyz+Wn04zwR3HZZXWU+QOARm1xK6WCUFesFCtr1gp8aPZ0QZjCFl8x5JPLEo0Bj5P 9wkJRIVQQZBUptDyLzT2Iql7McMJEKkhv+iFU4llIQDaJGTiuyQUTj2vp0QPfbTUJ3 jxo4Go8KMZz5Mnk3HA9JOSJZ+J8GsA/ldz1FLdJs= Date: Sun, 14 Jul 2019 18:14:28 +0100 From: Jonathan Cameron To: Martyn Welch Cc: Rob Herring , Mark Rutland , Hartmut Knaack , Lars-Peter Clausen , Peter Meerwald-Stadler , linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org, kernel@lists.collabora.co.uk, devicetree@vger.kernel.org, Sergei M Subject: Re: [PATCH v2 2/2] iio: light: noa1305: Add support for NOA1305 Message-ID: <20190714181428.5e6eb83c@archlinux> In-Reply-To: <20190628175713.14831-2-martyn.welch@collabora.com> References: <20190628175713.14831-1-martyn.welch@collabora.com> <20190628175713.14831-2-martyn.welch@collabora.com> X-Mailer: Claws Mail 3.17.3 (GTK+ 2.24.32; x86_64-pc-linux-gnu) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 28 Jun 2019 18:57:13 +0100 Martyn Welch wrote: > This driver adds the initial support for the ON Semiconductor > NOA1305 Ambient Light Sensor. >=20 > Originally written by Sergei Miroshnichenko. Found here: > https://github.com/EmcraftSystems/linux-upstream/commit/196d6cf897e632d= 2cb82d45484bd7a1bfdd5b6d9 >=20 > Signed-off-by: Sergei M > Signed-off-by: Martyn Welch A few minor things inline. Thanks, Jonathan > --- >=20 > Changes: > v2: Correcting authorship and SOB. >=20 > drivers/iio/light/Kconfig | 10 ++ > drivers/iio/light/Makefile | 1 + > drivers/iio/light/noa1305.c | 247 ++++++++++++++++++++++++++++++++++++ > 3 files changed, 258 insertions(+) > create mode 100644 drivers/iio/light/noa1305.c >=20 > diff --git a/drivers/iio/light/Kconfig b/drivers/iio/light/Kconfig > index 954c958cfc43..d1db0ec0d0f5 100644 > --- a/drivers/iio/light/Kconfig > +++ b/drivers/iio/light/Kconfig > @@ -309,6 +309,16 @@ config MAX44009 > To compile this driver as a module, choose M here: > the module will be called max44009. > =20 > +config NOA1305 > + tristate "ON Semiconductor NOA1305 ambient light sensor" > + depends on I2C > + help > + Say Y here if you want to build support for the ON Semiconductor > + NOA1305 ambient light sensor. > + > + To compile this driver as a module, choose M here: > + The module will be called noa1305. > + > config OPT3001 > tristate "Texas Instruments OPT3001 Light Sensor" > depends on I2C > diff --git a/drivers/iio/light/Makefile b/drivers/iio/light/Makefile > index e40794fbb435..00d1f9b98f39 100644 > --- a/drivers/iio/light/Makefile > +++ b/drivers/iio/light/Makefile > @@ -29,6 +29,7 @@ obj-$(CONFIG_LTR501) +=3D ltr501.o > obj-$(CONFIG_LV0104CS) +=3D lv0104cs.o > obj-$(CONFIG_MAX44000) +=3D max44000.o > obj-$(CONFIG_MAX44009) +=3D max44009.o > +obj-$(CONFIG_NOA1305) +=3D noa1305.o > obj-$(CONFIG_OPT3001) +=3D opt3001.o > obj-$(CONFIG_PA12203001) +=3D pa12203001.o > obj-$(CONFIG_RPR0521) +=3D rpr0521.o > diff --git a/drivers/iio/light/noa1305.c b/drivers/iio/light/noa1305.c > new file mode 100644 > index 000000000000..2c65c5c2e09a > --- /dev/null > +++ b/drivers/iio/light/noa1305.c > @@ -0,0 +1,247 @@ > +// SPDX-License-Identifier: GPL-2.0+ > +/* > + * Support for ON Semiconductor NOA1305 ambient light sensor > + * > + * Copyright (C) 2016 Emcraft Systems > + * Copyright (C) 2019 Collabora Ltd. > + */ > + > +#include > +#include > +#include > +#include > +#include > +#include > +#include > +#include > + > +#define NOA1305_REG_POWER_CONTROL 0x0 > +#define NOA1305_REG_RESET 0x1 > +#define NOA1305_REG_INTEGRATION_TIME 0x2 > +#define NOA1305_REG_INT_SELECT 0x3 > +#define NOA1305_REG_INT_THRESH_LSB 0x4 > +#define NOA1305_REG_INT_THRESH_MSB 0x5 > +#define NOA1305_REG_ALS_DATA_LSB 0x6 > +#define NOA1305_REG_ALS_DATA_MSB 0x7 > +#define NOA1305_REG_DEVICE_ID_LSB 0x8 > +#define NOA1305_REG_DEVICE_ID_MSB 0x9 > + > +#define NOA1305_DEVICE_ID 0x0519 > + > +#define NOA1305_POWER_ON 0x08 > +#define NOA1305_POWER_DOWN 0x00 > +#define NOA1305_RESET 0x10 It would be good to name these in a fashion that made it obvious they were values to write to the power control register. NOA1305_POWER_CONTROL_ON NOA1305_POWER_CONTROL_DOWN NOA1305_POWER_CONTROL_RESET perhaps? There are other conventions we tend to use in IIO such as #define NOA1305_POWER_CONTROL_REG 0x0 #define NOA1305_POWER_CONTROL_POWER_ON 0x08 etc.. =20 > + > +#define NOA1305_INT_ACTIVE_HIGH 0x01 > +#define NOA1305_INT_ACTIVE_LOW 0x02 > +#define NOA1305_INT_INACTIVE 0x03 > + > +#define NOA1305_INTEGR_TIME_800MS 0x00 > +#define NOA1305_INTEGR_TIME_400MS 0x01 > +#define NOA1305_INTEGR_TIME_200MS 0x02 > +#define NOA1305_INTEGR_TIME_100MS 0x03 > +#define NOA1305_INTEGR_TIME_50MS 0x04 > +#define NOA1305_INTEGR_TIME_25MS 0x05 > +#define NOA1305_INTEGR_TIME_12_5MS 0x06 > +#define NOA1305_INTEGR_TIME_6_25MS 0x07 > + > +#define NOA1305_DRIVER_NAME "noa1305" > + > +struct noa1305_priv { > + struct i2c_client *client; > + struct regmap *regmap; > + struct regulator *vin_reg; > +}; > + > +static int noa1305_measure(struct noa1305_priv *priv) > +{ > + u8 data[2]; > + int ret; > + > + ret =3D regmap_bulk_read(priv->regmap, NOA1305_REG_ALS_DATA_LSB, data, > + 2); > + if (ret < 0) > + return ret; > + > + return (data[1] << 8) | data[0]; Make data an explicit __le16 and use an endian swap here as on little endian platforms it'll be free. > +} > + > +static const struct iio_chan_spec noa1305_channels[] =3D { > + { > + .type =3D IIO_LIGHT, > + .info_mask_separate =3D BIT(IIO_CHAN_INFO_RAW), > + } > +}; > + > +static int noa1305_read_raw(struct iio_dev *indio_dev, > + struct iio_chan_spec const *chan, > + int *val, int *val2, long mask) > +{ > + int ret =3D -EINVAL; > + struct noa1305_priv *priv =3D iio_priv(indio_dev); > + > + switch (mask) { > + case IIO_CHAN_INFO_RAW: > + switch (chan->type) { > + case IIO_LIGHT: > + ret =3D noa1305_measure(priv); Hmm. So no information at all available on how this maps to illuminance? If the mapping is something you were thinking of adding later we would end up having to support this old ABI as well as the _PROCESSED version and, whilst it's happened occasionall= y, it isn't something we encourage. Of course, if there is no documented conversion function then we=20 can live with this interface! =46rom figure 6 in the datasheet it seems linear (unusual for a light sensor) so should be easy enough to do with offset and scale, in which case this is fine as it is. > + if (ret < 0) > + return ret; > + *val =3D ret; > + ret =3D IIO_VAL_INT; > + break; > + default: > + break; > + } > + break; > + default: > + break; > + } > + > + return ret; > +} > + > +static const struct iio_info noa1305_info =3D { > + .read_raw =3D noa1305_read_raw, > +}; > + > +static bool noa1305_writable_reg(struct device *dev, unsigned int reg) > +{ > + switch (reg) { > + case NOA1305_REG_POWER_CONTROL: > + case NOA1305_REG_RESET: > + case NOA1305_REG_INTEGRATION_TIME: > + case NOA1305_REG_INT_SELECT: > + case NOA1305_REG_INT_THRESH_LSB: > + case NOA1305_REG_INT_THRESH_MSB: > + return true; > + default: > + return false; > + } > +} > + > +static const struct regmap_config noa1305_regmap_config =3D { > + .name =3D NOA1305_DRIVER_NAME, > + .reg_bits =3D 8, > + .val_bits =3D 8, > + .max_register =3D NOA1305_REG_DEVICE_ID_MSB, > + .writeable_reg =3D noa1305_writable_reg, > +}; > + > +static int noa1305_probe(struct i2c_client *client, > + const struct i2c_device_id *id) > +{ > + struct noa1305_priv *priv; > + struct iio_dev *indio_dev; > + struct regmap *regmap; > + u8 data[2]; > + unsigned int dev_id =3D 0; Always set in any path where it is used, so don't do it here. > + int ret; > + > + indio_dev =3D devm_iio_device_alloc(&client->dev, sizeof(*priv)); > + if (!indio_dev) > + return -ENOMEM; > + > + regmap =3D devm_regmap_init_i2c(client, &noa1305_regmap_config); > + if (IS_ERR(regmap)) { > + dev_err(&client->dev, "Regmap initialization failed.\n"); > + return PTR_ERR(regmap); > + } > + > + priv =3D iio_priv(indio_dev); > + > + priv->vin_reg =3D devm_regulator_get(&client->dev, "vin"); > + if (IS_ERR(priv->vin_reg)) { > + dev_err(&client->dev, "get regulator vin failed\n"); > + return PTR_ERR(priv->vin_reg); > + } > + > + ret =3D regulator_enable(priv->vin_reg); > + if (ret) { > + dev_err(&client->dev, "enable regulator vin failed\n"); > + return ret; > + } > + > + i2c_set_clientdata(client, indio_dev); > + priv->client =3D client; > + priv->regmap =3D regmap; > + > + ret =3D regmap_bulk_read(regmap, NOA1305_REG_DEVICE_ID_LSB, data, 2); > + if (ret < 0) { > + dev_err(&client->dev, "ID reading failed: %d\n", ret); > + goto error_disable_reg; > + } > + > + dev_id =3D (data[1] << 8) | data[0]; This is just unwinding endianness, use the __*e16_to_cpu function and an appropriate type instead of u8[2] for this. > + if (dev_id !=3D NOA1305_DEVICE_ID) { > + dev_err(&client->dev, "Unknown device ID: 0x%x\n", dev_id); > + ret =3D -ENODEV; > + goto error_disable_reg; > + } > + > + regmap_write(regmap, NOA1305_REG_POWER_CONTROL, NOA1305_POWER_ON); I2C buses aren't exactly the most reliable things out there, so I'd expect to see error handling on all these writes. > + regmap_write(regmap, NOA1305_REG_RESET, NOA1305_RESET); > + regmap_write(regmap, NOA1305_REG_INTEGRATION_TIME, > + NOA1305_INTEGR_TIME_800MS); > + regmap_write(regmap, NOA1305_REG_INT_SELECT, NOA1305_INT_INACTIVE); > + > + indio_dev->dev.parent =3D &client->dev; > + indio_dev->info =3D &noa1305_info; > + indio_dev->channels =3D noa1305_channels; > + indio_dev->num_channels =3D ARRAY_SIZE(noa1305_channels); > + indio_dev->name =3D NOA1305_DRIVER_NAME; > + indio_dev->modes =3D INDIO_DIRECT_MODE; > + > + ret =3D devm_iio_device_register(&client->dev, indio_dev); There is a race here because the regulator disable will happen in remove before we remove the userspace interfaces. Hence it is possible to try to read from the device whilst it has no power. Two options. Either stop using devm for everything after the regulator enable, or use a devm_add_action_or_reset to make the regulator disable occur in the devm managed sequence rather than in the manual remove funciton. > + if (ret) { > + dev_err(&client->dev, "registering device failed\n"); > + goto error_disable_reg; > + } > + > + return ret; > + > +error_disable_reg: > + regulator_disable(priv->vin_reg); > + > + return ret; > +} > + > +static int noa1305_remove(struct i2c_client *client) > +{ > + struct iio_dev *indio_dev =3D i2c_get_clientdata(client); > + struct noa1305_priv *priv =3D iio_priv(indio_dev); > + > + regulator_disable(priv->vin_reg); > + > + return 0; > +} > + > +static const struct of_device_id noa1305_of_match[] =3D { > + { .compatible =3D "onnn,noa1305" }, > + { } > +}; > +MODULE_DEVICE_TABLE(of, noa1305_of_match); > + > +static const struct i2c_device_id noa1305_ids[] =3D { > + { "noa1305", 0 }, > + { } > +}; > +MODULE_DEVICE_TABLE(i2c, noa1305_id); > + nitpick: One line is plenty. > + > +static struct i2c_driver noa1305_driver =3D { > + .driver =3D { > + .name =3D NOA1305_DRIVER_NAME, > + .of_match_table =3D noa1305_of_match, > + }, > + .probe =3D noa1305_probe, > + .remove =3D noa1305_remove, > + .id_table =3D noa1305_ids, > +}; > + > +module_i2c_driver(noa1305_driver); > + > +MODULE_AUTHOR("Sergei Miroshnichenko "); > +MODULE_AUTHOR("Martyn Welch +MODULE_DESCRIPTION("ON Semiconductor NOA1305 ambient light sensor"); > +MODULE_LICENSE("GPL");