| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Sat, 08 Sep 2012 11:25:25 +0100
From: Jonathan Cameron <jic23@...nel.org>
To: "Kim, Milo" <Milo.Kim@...com>
CC: Jonathan Cameron <jic23@....ac.uk>,
Lars-Peter Clausen <lars@...afoo.de>,
"linux-iio@...r.kernel.org" <linux-iio@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v4] iio: adc: add new lp8788 adc driver
On 09/07/2012 09:51 AM, Kim, Milo wrote:
Mostly looks fine. A small confusion has occured with the consumer
interface though and the read_raw needs a bit of reorganizing.
Nearly there!
> TI LP8788 PMU provides regulators, battery charger, ADC,
> RTC, backlight driver and current sinks.
>
> This patch enables the LP8788 ADC functions.
>
> The LP8788 ADC has 13 ADC input selection and supports 12-bit resolution.
> Internal operation of getting ADC is access to registers of LP8788.
> The LP8788 ADC uses exported functions for accessing these registers.
> (exported by LP8788 MFD device driver)
>
> This driver supports IIO_CHAN_INFO_RAW and SCALE.
> The scale is micro unit.
> So the IIO consumer can get the result as following.
>
> result = ADC raw value * (scale_int + scale_part / 1000000)
>
>
> (The IIO map for the IIO consumer)
I think there is some confusion about who the consumer is. With
iio_map structures the consumer is the driver that is 'using' the
iio driver to read from a device. Hence this driver shouldn't know
anything about the consumer side of things and that should always
be passed in from platform data.
>
> The ADC input can be selective in the platform side.
> Even though this platform data is not defined,
> the default IIO map is created for supporting the power supply driver.
> The battery voltage and temperature are used inside this driver.
Ah, I guess this may make sense in a device as heaving tied as this.
however,
>
> (History)
>
> Patch v4.
> Fix adc_raw function: support RAW and SCALE channel info.
> Change LP8788 ADC platform data - iio map.
> Enables the default IIO map.
>
> Patch v3.
> Fix wrong size of allocating iio private data.
> Fix coding styles.
>
> Patch v2.
> Support RAW and SCALE interface for IIO consumer.
> Clean up the iio channel spec macro.
>
> Signed-off-by: Milo(Woogyom) Kim <milo.kim@...com>
> ---
> drivers/iio/adc/Kconfig | 6 ++
> drivers/iio/adc/Makefile | 1 +
> drivers/iio/adc/lp8788_adc.c | 242 ++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 249 insertions(+)
> 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..33d9aaf
> --- /dev/null
> +++ b/drivers/iio/adc/lp8788_adc.c
> @@ -0,0 +1,242 @@
> +/*
> + * TI LP8788 MFD - ADC driver
> + *
> + * Copyright 2012 Texas Instruments
> + *
> + * Author: Milo(Woogyom) Kim <milo.kim@...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/delay.h>
> +#include <linux/iio/iio.h>
> +#include <linux/iio/driver.h>
> +#include <linux/iio/machine.h>
> +#include <linux/mfd/lp8788.h>
> +#include <linux/module.h>
> +#include <linux/platform_device.h>
> +#include <linux/slab.h>
> +
> +/* register address */
> +#define LP8788_ADC_CONF 0x60
> +#define LP8788_ADC_RAW 0x61
> +#define LP8788_ADC_DONE 0x63
> +
> +#define ADC_CONV_START 1
> +
> +struct lp8788_adc {
> + struct lp8788 *lp;
> + struct iio_map *map;
> +};
> +
> +static const int lp8788_scale[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 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, result;
> + u8 data, rawdata[2];
> + int size = ARRAY_SIZE(rawdata);
> + enum lp8788_adc_id id = chan->channel;
> +
Sorry, didn't notice this before but you probably want some
locking around the write / read pair in here as there is nothing
stopping simultaneous reads of different channels.
> + data = (id << 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--) {
> + usleep_range(100, 200);
> +
> + 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;
> +
> + msb = (rawdata[0] << 4) & 0x00000ff0;
> + lsb = (rawdata[1] >> 4) & 0x0000000f;
> + result = msb | lsb;
> +
> + switch (mask) {
> + case IIO_CHAN_INFO_RAW:
> + *val = result;
> + return IIO_VAL_INT;
Why are you reading from the adc for the scale case?
A bit of reordering in this function is needed to avoid that.
> + case IIO_CHAN_INFO_SCALE:
> + *val = (lp8788_scale[id] / 1000) * 1000;
> + *val2 = (lp8788_scale[id] % 1000) * 1000000;
> + return IIO_VAL_INT_PLUS_MICRO;
> + default:
> + break;
> + }
> +
> +err:
> + return -EINVAL;
> +}
> +
> +static const struct iio_info lp8788_adc_info = {
> + .read_raw = &lp8788_adc_read_raw,
> + .driver_module = THIS_MODULE,
> +};
> +
> +#define LP8788_CHAN(_id, _type) { \
> + .type = _type, \
> + .indexed = 1, \
> + .channel = LPADC_##_id, \
> + .info_mask = IIO_CHAN_INFO_RAW_SEPARATE_BIT | \
> + IIO_CHAN_INFO_SCALE_SEPARATE_BIT, \
> + .address = LP8788_ADC_RAW, \
> + .scan_type = IIO_ST('u', 12, 16, 0), \
> + .scan_index = 1, \
> + .datasheet_name = #_id, \
> +}
> +
> +static const struct iio_chan_spec lp8788_adc_channels[] = {
> + [LPADC_VBATT_5P5] = LP8788_CHAN(VBATT_5P5, IIO_VOLTAGE),
> + [LPADC_VIN_CHG] = LP8788_CHAN(VIN_CHG, IIO_VOLTAGE),
> + [LPADC_IBATT] = LP8788_CHAN(IBATT, IIO_CURRENT),
> + [LPADC_IC_TEMP] = LP8788_CHAN(IC_TEMP, IIO_TEMP),
> + [LPADC_VBATT_6P0] = LP8788_CHAN(VBATT_6P0, IIO_VOLTAGE),
> + [LPADC_VBATT_5P0] = LP8788_CHAN(VBATT_5P0, IIO_VOLTAGE),
> + [LPADC_ADC1] = LP8788_CHAN(ADC1, IIO_VOLTAGE),
> + [LPADC_ADC2] = LP8788_CHAN(ADC2, IIO_VOLTAGE),
> + [LPADC_VDD] = LP8788_CHAN(VDD, IIO_VOLTAGE),
> + [LPADC_VCOIN] = LP8788_CHAN(VCOIN, IIO_VOLTAGE),
> + [LPADC_VDD_LDO] = LP8788_CHAN(VDD_LDO, IIO_VOLTAGE),
> + [LPADC_ADC3] = LP8788_CHAN(ADC3, IIO_VOLTAGE),
> + [LPADC_ADC4] = LP8788_CHAN(ADC4, IIO_VOLTAGE),
> +};
> +
> +/* default maps used by iio consumer (lp8788-charger driver) */
Looking quickly at the code in inkern.c I'm not entirely sure why
it doesn't specify that the iio_map is constant (and hence allow
you to have it constant here...) I'll fix that at some point if
someone else doesn't get to it first...
> +static struct iio_map lp8788_default_iio_maps[] = {
> + {
Isn't this backwards. This driver is the lp8788-adc why would
it be consuming itself. You state in the comments that these
go to the powersupply driver hence it consumer_dev_name
should be set to the name of that devices rather than this one.
> + .consumer_dev_name = "lp8788-adc",
> + .consumer_channel = "vbatt-5p0",
> + .adc_channel_label = "VBATT_5P0",
> + },
> + {
> + .consumer_dev_name = "lp8788-adc",
> + .consumer_channel = "adc1",
> + .adc_channel_label = "ADC1",
> + },
> + { }
> +};
> +
> +static int lp8788_iio_map_register(struct iio_dev *indio_dev,
> + struct lp8788_platform_data *pdata,
> + struct lp8788_adc *adc)
> +{
> + struct iio_map *map;
> + int ret;
> +
> + map = !pdata || !pdata->adc_pdata ?
> + lp8788_default_iio_maps : pdata->adc_pdata;
> +
> + ret = iio_map_array_register(indio_dev, map);
> + if (ret) {
> + dev_err(adc->lp->dev, "iio map err: %d\n", ret);
> + return ret;
> + }
> +
> + adc->map = map;
> + return 0;
> +}
> +
> +static void lp8788_iio_map_unregister(struct iio_dev *indio_dev,
> + struct lp8788_adc *adc)
> +{
> + if (adc->map)
> + iio_map_array_unregister(indio_dev, adc->map);
> +}
> +
> +static int __devinit lp8788_adc_probe(struct platform_device *pdev)
> +{
> + struct lp8788 *lp = dev_get_drvdata(pdev->dev.parent);
> + struct iio_dev *indio_dev;
> + struct lp8788_adc *adc;
> + int ret;
> +
> + indio_dev = iio_device_alloc(sizeof(*adc));
> + if (!indio_dev)
> + return -ENOMEM;
> +
> + adc = iio_priv(indio_dev);
> + adc->lp = lp;
> + platform_set_drvdata(pdev, indio_dev);
> +
> + ret = lp8788_iio_map_register(indio_dev, lp->pdata, adc);
> + if (ret)
> + goto err_iio_map;
> +
> + 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_device;
> + }
> +
> + return 0;
> +
> +err_iio_device:
> + lp8788_iio_map_unregister(indio_dev, adc);
> +err_iio_map:
> + iio_device_free(indio_dev);
> + return ret;
> +}
> +
> +static int __devexit lp8788_adc_remove(struct platform_device *pdev)
> +{
> + struct iio_dev *indio_dev = platform_get_drvdata(pdev);
> + struct lp8788_adc *adc = iio_priv(indio_dev);
> +
> + iio_device_unregister(indio_dev);
> + lp8788_iio_map_unregister(indio_dev, adc);
> + 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");
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists