All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jonathan Cameron <jic23@kernel.org>
To: "Kim, Milo" <Milo.Kim@ti.com>
Cc: "jic23@cam.ac.uk" <jic23@cam.ac.uk>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-iio@vger.kernel.org" <linux-iio@vger.kernel.org>,
	"sameo@linux.intel.com" <sameo@linux.intel.com>
Subject: Re: [PATCH 2/3] iio: adc: add new lp8788 adc driver
Date: Thu, 09 Aug 2012 11:15:21 +0100	[thread overview]
Message-ID: <50238DB9.6040709@kernel.org> (raw)
In-Reply-To: <A874F61F95741C4A9BA573A70FE3998F41EEBC4F@DBDE02.ent.ti.com>

On 08/09/2012 09:22 AM, Kim, Milo wrote:
> TI LP8788 has ADC function.
> The result of LP878 ADC is used in the LP8788 power supply driver.
> (such like getting the battery voltage, temperature and etc)
> 
Hi,

This is mostly fine though things have gotten a little confused
wrt to the handling iio_priv in the probe and remove so that
needs cleaning up.  A few other minor bits inline.

Thanks,

Jonathan
> Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
> ---
>  drivers/iio/adc/Kconfig      |    6 +
>  drivers/iio/adc/Makefile     |    1 +
>  drivers/iio/adc/lp8788_adc.c |  240 ++++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 247 insertions(+), 0 deletions(-)
>  create mode 100644 drivers/iio/adc/lp8788_adc.c
> 
> diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
> index 8a78b4f..30c06ed 100644
> --- a/drivers/iio/adc/Kconfig
> +++ b/drivers/iio/adc/Kconfig
> @@ -22,4 +22,10 @@ config AT91_ADC
>  	help
>  	  Say yes here to build support for Atmel AT91 ADC.
>  
> +config LP8788_ADC
> +	bool "LP8788 ADC driver"
> +	depends on MFD_LP8788
> +	help
> +	  Say yes here to build support for TI LP8788 ADC.
> +
>  endmenu
> diff --git a/drivers/iio/adc/Makefile b/drivers/iio/adc/Makefile
> index 52eec25..72f9a76 100644
> --- a/drivers/iio/adc/Makefile
> +++ b/drivers/iio/adc/Makefile
> @@ -4,3 +4,4 @@
>  
>  obj-$(CONFIG_AD7266) += ad7266.o
>  obj-$(CONFIG_AT91_ADC) += at91_adc.o
> +obj-$(CONFIG_LP8788_ADC) += lp8788_adc.o
> diff --git a/drivers/iio/adc/lp8788_adc.c b/drivers/iio/adc/lp8788_adc.c
> new file mode 100644
> index 0000000..30767d5
> --- /dev/null
> +++ b/drivers/iio/adc/lp8788_adc.c
> @@ -0,0 +1,240 @@
> +/*
> + * TI LP8788 MFD - ADC driver
> + *
> + * Copyright 2012 Texas Instruments
> + *
> + * Author: Milo(Woogyom) Kim <milo.kim@ti.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + *
> + */
> +
> +#include <linux/module.h>
> +#include <linux/slab.h>
> +#include <linux/err.h>
> +#include <linux/platform_device.h>
> +#include <linux/delay.h>
> +#include <linux/iio/iio.h>
> +#include <linux/iio/driver.h>
> +#include <linux/mfd/lp8788.h>
> +
> +/* register address */
> +#define LP8788_ADC_CONF			0x60
> +#define LP8788_ADC_RAW			0x61
> +#define LP8788_ADC_DONE			0x63
> +
> +#define START_ADC_CHANNEL		LPADC_VBATT_5P5
> +#define END_ADC_CHANNEL			LPADC_MAX
> +#define ADC_CONV_START			1
> +#define ADC_CONV_DELAY_US		100
> +
> +struct lp8788_adc {
> +	struct lp8788 *lp;

As explained below I would drop this pointer.
The iio_priv with the suggested changes just stores a
pointer to lp.
> +	struct iio_dev *indio_dev;
> +};
> +
> +static const int adc_const[LPADC_MAX] = {
> +	[LPADC_VBATT_5P5] = 1343,
> +	[LPADC_VIN_CHG]   = 3052,
> +	[LPADC_IBATT]     = 610,
> +	[LPADC_IC_TEMP]   = 610,
> +	[LPADC_VBATT_6P0] = 1465,
> +	[LPADC_VBATT_5P0] = 1221,
> +	[LPADC_ADC1]      = 610,
> +	[LPADC_ADC2]      = 610,
> +	[LPADC_VDD]       = 1025,
> +	[LPADC_VCOIN]     = 757,
> +	[LPADC_VDD_LDO]   = 610,
> +	[LPADC_ADC3]      = 610,
> +	[LPADC_ADC4]      = 610,
> +};
> +
> +static inline unsigned int _get_adc_micro_unit(enum lp8788_adc_id id,
> +					unsigned int adc_result)
> +{
> +	return adc_const[id] * ((adc_result * 1000 + 500) / 1000);
> +}
You haven't marked the channels as processed (e.g. converted into correct units)
so I would do these as chan_info elements (scale and offset)
> +
> +static int lp8788_adc_read_raw(struct iio_dev *indio_dev,
> +			struct iio_chan_spec const *chan,
> +			int *val, int *val2, long mask)
> +{
> +	struct lp8788_adc *adc = iio_priv(indio_dev);
> +	int retry = 5;
> +	unsigned int msb, lsb;
> +	u8 data, rawdata[2], shift;
> +	int size = ARRAY_SIZE(rawdata);
only used in one place, so do it inline.

> +
> +	if (mask != IIO_CHAN_INFO_RAW)
> +		return -EINVAL;
> +
> +	data = (chan->channel << 1) | ADC_CONV_START;
> +	if (lp8788_write_byte(adc->lp, LP8788_ADC_CONF, data))
> +		goto err;
> +
> +	/* retry until adc conversion is done */
> +	data = 0;
> +	while (retry--) {
> +		udelay(ADC_CONV_DELAY_US);
> +
> +		if (lp8788_read_byte(adc->lp, LP8788_ADC_DONE, &data))
> +			goto err;
> +
> +		/* conversion done */
> +		if (data)
> +			break;
> +	}
> +
> +	if (lp8788_read_multi_bytes(adc->lp, LP8788_ADC_RAW, rawdata, size))
> +		goto err;
> +
This is a little odd. You establish the shift from the channel type, but
then have the masks hard coded. I'd just hard code the lot to
make it slightly easier to read

> +	shift = chan->scan_type.shift;
> +	msb = (rawdata[0] << shift) & 0x00000ff0;
> +	lsb = (rawdata[1] >> shift) & 0x0000000f;
> +	*val = _get_adc_micro_unit(chan->channel, msb | lsb);
> +
> +	return IIO_VAL_INT;
> +err:
> +	return -EINVAL;
> +}
> +
> +static const struct iio_info lp8788_adc_info = {
> +	.read_raw = &lp8788_adc_read_raw,
> +	.driver_module = THIS_MODULE,
> +};
> +
> +#define LP8788_COMMON_CH(id)					\
> +		.indexed = 1,					\
> +		.output = 1,					\
err. these aren't all outputs... Actually I'm not sure any of them are...
> +		.channel = LPADC_##id,				\
> +		.info_mask = IIO_CHAN_INFO_RAW_SEPARATE_BIT,	\
> +		.address = LP8788_ADC_RAW,			\
> +		.scan_type = IIO_ST('u', 8, 12, 4),		\
> +		.scan_index = 1,				\
> +		.datasheet_name = #id,				\
Given how you are using this macro I'd just make it take an additional
parameter then you can effectively drop all the ones below.
So have
#define LP8788_CHAN(id, type)

> +
> +#define LP8788_V_CHAN(id)	{				\
> +		.type = IIO_VOLTAGE,				\
> +		LP8788_COMMON_CH(id)				\
> +}
> +
> +#define LP8788_I_CHAN(id)	{				\
> +		.type = IIO_CURRENT,				\
> +		LP8788_COMMON_CH(id)				\
> +}
> +
> +#define LP8788_T_CHAN(id)	{				\
> +		.type = IIO_TEMP,				\
> +		LP8788_COMMON_CH(id)				\
> +}
> +
> +static struct iio_chan_spec lp8788_adc_channels[] = {
> +	[LPADC_VBATT_5P5] = LP8788_V_CHAN(VBATT_5P5),
> +	[LPADC_VIN_CHG]   = LP8788_V_CHAN(VIN_CHG),
> +	[LPADC_IBATT]     = LP8788_I_CHAN(IBATT),
> +	[LPADC_IC_TEMP]   = LP8788_T_CHAN(IC_TEMP),
> +	[LPADC_VBATT_6P0] = LP8788_V_CHAN(VBATT_6P0),
> +	[LPADC_VBATT_5P0] = LP8788_V_CHAN(VBATT_5P0),
> +	[LPADC_ADC1]      = LP8788_V_CHAN(ADC1),
> +	[LPADC_ADC2]      = LP8788_V_CHAN(ADC2),
> +	[LPADC_VDD]       = LP8788_V_CHAN(VDD),
> +	[LPADC_VCOIN]     = LP8788_V_CHAN(VCOIN),
> +	[LPADC_VDD_LDO]   = LP8788_V_CHAN(VDD_LDO),
> +	[LPADC_ADC3]      = LP8788_V_CHAN(ADC3),
> +	[LPADC_ADC4]      = LP8788_V_CHAN(ADC4),
> +};
> +
> +static int __devinit lp8788_adc_probe(struct platform_device *pdev)
> +{
> +	struct lp8788 *lp = dev_get_drvdata(pdev->dev.parent);
> +	struct iio_map *map;
> +	struct lp8788_adc *adc;
> +	struct iio_dev *indio_dev;
> +	int i, ret;
> +
> +	adc = devm_kzalloc(lp->dev, sizeof(struct lp8788_adc), GFP_KERNEL);
> +	if (!adc)
> +		return -ENOMEM;
> +
This has me confused. The parameter is for private data (e.g. lp8788_adc here).
It already allocates enough space for a struct iio_dev.   If you want to do
it as here, you'll need to have the parameter as 0.
> +	indio_dev = iio_device_alloc(sizeof(struct iio_dev));
> +	if (!indio_dev)
> +		return -ENOMEM;
> +
> +	adc = iio_priv(indio_dev);
That's just set it to point at a region sized for a struct iio_dev.

What I think you meant to do was drop the adc = devm_kzalloc
and do
indio_dev = iio_device_alloc(sizeof(*adc));
> +	adc->lp = lp;
> +	adc->indio_dev = indio_dev;
As a general rule we avoid this sort of reflection back. It's normally
not necessary given the indio_dev should almost always be available...

> +	platform_set_drvdata(pdev, adc);
set that to be the indio_dev rather than adc and you won't need the
adc->indio_dev pointer.
> +
> +	if (lp->pdata) {
> +		for (i = START_ADC_CHANNEL; i < END_ADC_CHANNEL ; i++) {
> +			map = lp->pdata->adc_pdata[i];
> +			ret = iio_map_array_register(indio_dev, map);
> +			if (ret) {
> +				dev_err(lp->dev, "iio map err: %d\n", ret);
> +				goto err_iio_map_array;
> +			}
> +		}
> +	}
> +
> +	indio_dev->dev.parent = lp->dev;
> +	indio_dev->name = pdev->name;
> +	indio_dev->modes = INDIO_DIRECT_MODE;
> +	indio_dev->info = &lp8788_adc_info;
> +	indio_dev->channels = lp8788_adc_channels;
> +	indio_dev->num_channels = ARRAY_SIZE(lp8788_adc_channels);
> +
> +	ret = iio_device_register(indio_dev);
> +	if (ret) {
> +		dev_err(lp->dev, "iio dev register err: %d\n", ret);
> +		goto err_iio_register;
> +	}
> +
> +	return 0;
> +
> +err_iio_map_array:
> +	while (--i >= START_ADC_CHANNEL) {
> +		map = lp->pdata->adc_pdata[i];
> +		iio_map_array_unregister(indio_dev, map);
> +	}
> +err_iio_register:
> +	iio_device_free(indio_dev);
> +	return ret;
> +}
> +
> +static int __devexit lp8788_adc_remove(struct platform_device *pdev)
> +{
> +	struct lp8788_adc *adc = platform_get_drvdata(pdev);
> +	struct lp8788 *lp = adc->lp;
if the platform data is set to the indio_dev this isn't needed and
adc can retrieved via iio_priv.
> +	struct iio_dev *indio_dev = adc->indio_dev;
> +	struct iio_map *map;
> +	int i;
> +
> +	if (lp->pdata) {
> +		for (i = START_ADC_CHANNEL; i < END_ADC_CHANNEL ; i++) {
> +			map = lp->pdata->adc_pdata[i];
> +			iio_map_array_unregister(indio_dev, map);
> +		}
> +	}
> +
> +	iio_device_unregister(indio_dev);
> +	iio_device_free(indio_dev);
> +
> +	return 0;
> +}
> +
> +static struct platform_driver lp8788_adc_driver = {
> +	.probe = lp8788_adc_probe,
> +	.remove = __devexit_p(lp8788_adc_remove),
> +	.driver = {
> +		.name = LP8788_DEV_ADC,
> +		.owner = THIS_MODULE,
> +	},
> +};
> +module_platform_driver(lp8788_adc_driver);
> +
> +MODULE_DESCRIPTION("Texas Instruments LP8788 ADC Driver");
> +MODULE_AUTHOR("Milo Kim");
> +MODULE_LICENSE("GPL");
> +MODULE_ALIAS("platform:lp8788-adc");
> 

  reply	other threads:[~2012-08-09 10:15 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-08-09  8:22 [PATCH 2/3] iio: adc: add new lp8788 adc driver Kim, Milo
2012-08-09  8:22 ` Kim, Milo
2012-08-09 10:15 ` Jonathan Cameron [this message]
2012-08-10  7:07   ` Kim, Milo
2012-08-10  7:07     ` Kim, Milo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=50238DB9.6040709@kernel.org \
    --to=jic23@kernel.org \
    --cc=Milo.Kim@ti.com \
    --cc=jic23@cam.ac.uk \
    --cc=linux-iio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sameo@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.