Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Sat, 21 Nov 2020 17:56:29 +0000
From:   Jonathan Cameron <jic23@kernel.org>
To:     Ye Xiang <xiang.ye@intel.com>
Cc:     jikos@kernel.org, srinivas.pandruvada@linux.intel.com,
        linux-input@vger.kernel.org, linux-iio@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 4/4] iio: hid-sensors: Add hinge sensor driver
Message-ID: <20201121175629.057031af@archlinux>
In-Reply-To: <20201119100331.2594-5-xiang.ye@intel.com>
References: <20201119100331.2594-1-xiang.ye@intel.com>
        <20201119100331.2594-5-xiang.ye@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Thu, 19 Nov 2020 18:03:31 +0800
Ye Xiang <xiang.ye@intel.com> wrote:

> The Hinge sensor is a common custom sensor on laptops. It calculates
> the angle between the lid (screen) and the base (keyboard). In addition,
> it also exposes screen and the keyboard angels with respect to the
> ground. Applications can easily get laptop's status in space through
> this sensor, in order to display appropriate user interface.

I'm a little unclear on why the 3 axes aren't treated as a single sensor.
You seem to always grab the 3 together or am I missing something?

That will greatly simplify things and get rid of the need to have
a shared trigger with the problems that causes in the previous
patch.

Thanks,

Jonathan

> 
> Signed-off-by: Ye Xiang <xiang.ye@intel.com>
> ---
>  .../hid-sensors/hid-sensor-attributes.c       |   2 +
>  drivers/iio/position/Kconfig                  |  16 +
>  drivers/iio/position/Makefile                 |   3 +
>  .../iio/position/hid-sensor-custom-hinge.c    | 412 ++++++++++++++++++

Given it's custom probably needs a more specific name.  I guess
hid-sensor-custom-intel-hinge.c might be safe?

Same for other places we need names in here.

>  4 files changed, 433 insertions(+)
>  create mode 100644 drivers/iio/position/hid-sensor-custom-hinge.c
> 
> diff --git a/drivers/iio/common/hid-sensors/hid-sensor-attributes.c b/drivers/iio/common/hid-sensors/hid-sensor-attributes.c
> index 442ff787f7af..5b822a4298a0 100644
> --- a/drivers/iio/common/hid-sensors/hid-sensor-attributes.c
> +++ b/drivers/iio/common/hid-sensors/hid-sensor-attributes.c
> @@ -71,6 +71,8 @@ static struct {
>  	{HID_USAGE_SENSOR_TEMPERATURE, HID_USAGE_SENSOR_UNITS_DEGREES, 1000, 0},
>  
>  	{HID_USAGE_SENSOR_HUMIDITY, 0, 1000, 0},
> +	{HID_USAGE_SENSOR_HINGE, 0, 0, 17453293},
> +	{HID_USAGE_SENSOR_HINGE, HID_USAGE_SENSOR_UNITS_DEGREES, 0, 17453293},
>  };
>  
>  static void simple_div(int dividend, int divisor, int *whole,
> diff --git a/drivers/iio/position/Kconfig b/drivers/iio/position/Kconfig
> index eda67f008c5b..0346f6f2b422 100644
> --- a/drivers/iio/position/Kconfig
> +++ b/drivers/iio/position/Kconfig
> @@ -16,4 +16,20 @@ config IQS624_POS
>  	  To compile this driver as a module, choose M here: the module
>  	  will be called iqs624-pos.
>  
> +config HID_SENSOR_CUSTOM_HINGE
> +	depends on HID_SENSOR_HUB
> +	select IIO_BUFFER
> +	select IIO_TRIGGERED_BUFFER
> +	select HID_SENSOR_IIO_COMMON
> +	select HID_SENSOR_IIO_TRIGGER
> +	tristate "HID Hinge"
> +	help
> +	  This sensor present three angles, hinge angel, screen angles
> +	  and keyboard angle respect to horizon (ground).
> +	  Say yes here to build support for the HID SENSOR CUSTOM
> +	  HINGE.

Capitalization is a bit odd looking. I'd drop it.

> +
> +	  To compile this driver as a module, choose M here: the
> +	  module will be called hid-sensor-custom-hinge.
> +
>  endmenu
> diff --git a/drivers/iio/position/Makefile b/drivers/iio/position/Makefile
> index 3cbe7a734352..7a6225977a01 100644
> --- a/drivers/iio/position/Makefile
> +++ b/drivers/iio/position/Makefile
> @@ -5,3 +5,6 @@
>  # When adding new entries keep the list in alphabetical order
>  
>  obj-$(CONFIG_IQS624_POS)	+= iqs624-pos.o
> +
> +obj-$(CONFIG_HID_SENSOR_CUSTOM_HINGE) += hid-sensor-custom-hinge.o

Alphabetical order preferred.

> +ccflags-y	+= -I$(srctree)/drivers/iio/common/hid-sensors

Why?

> diff --git a/drivers/iio/position/hid-sensor-custom-hinge.c b/drivers/iio/position/hid-sensor-custom-hinge.c
> new file mode 100644
> index 000000000000..a91b333f36fa
> --- /dev/null
> +++ b/drivers/iio/position/hid-sensor-custom-hinge.c
> @@ -0,0 +1,412 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * HID Sensors Driver
> + * Copyright (c) 2020, Intel Corporation.
> + */
> +#include <linux/hid-sensor-hub.h>
> +#include <linux/iio/buffer.h>
> +#include <linux/iio/iio.h>
> +#include <linux/platform_device.h>
> +
> +#include "hid-sensor-trigger.h"
> +
> +/* Channel definitions */
> +static const struct iio_chan_spec hinge_channels[] = {
> +	{ .type = IIO_ANGL,
> +	  .info_mask_separate = BIT(IIO_CHAN_INFO_RAW),
> +	  .info_mask_shared_by_type =
> +		  BIT(IIO_CHAN_INFO_OFFSET) | BIT(IIO_CHAN_INFO_SCALE) |
> +		  BIT(IIO_CHAN_INFO_SAMP_FREQ) | BIT(IIO_CHAN_INFO_HYSTERESIS),
> +	  .scan_type.realbits = 16,
> +	  .scan_type.storagebits = 32,

It a bit odd to see a single channel that is 16 bits inside a 32 bit with
no shift or similar.  Why not just pack it into 16 bits?

> +	  .scan_type.sign = 's',
> +	  .scan_index = 0 },
> +
> +	IIO_CHAN_SOFT_TIMESTAMP(1)
> +};
> +
> +struct hinge_state {
> +	struct iio_dev *indio_dev;
> +	struct hid_sensor_hub_attribute_info hinge;
> +	/* Reserve for 1 channel + pading + timestamp */
> +	u32 hinge_val[1 + 3];

__aligned(8)

see below for requirements on this.
Perhaps better to use

	struct hinge_scan {
		u32 val;
		s64 timestamp __aligned(8); // Note this is needed for x86_32
	} scan;

> +	int scale_pre_decml;
> +	int scale_post_decml;
> +	int scale_precision;
> +	int value_offset;
> +	int64_t timestamp;
> +	u32 hinge_address;
> +};
> +
> +#define IIO_DEV_NUM 3

That needs a prefix to make it clear it's not a generic constant
but is specific to this driver.

> +
> +struct hinge_group {
> +	struct hinge_state *hg_states[IIO_DEV_NUM];
> +	struct hid_sensor_hub_callbacks callbacks;
> +	struct hid_sensor_common common_attributes;
> +};
> +
> +static struct hinge_group *hg_group;

We shouldn't see globals like this. Please figure out how to avoid it.

> +
> +/* Channel read_raw handler */
> +static int hinge_read_raw(struct iio_dev *indio_dev,
> +			  struct iio_chan_spec const *chan, int *val, int *val2,
> +			  long mask)
> +{
> +	struct hinge_state *hg_state = iio_priv(indio_dev);
> +	struct hid_sensor_hub_device *hsdev;
> +	int report_id = -1;
> +	int ret_type;
> +	s32 min;
> +
> +	hsdev = hg_group->common_attributes.hsdev;
> +
> +	*val = 0;
> +	*val2 = 0;
> +	switch (mask) {
> +	case IIO_CHAN_INFO_RAW:
> +		hid_sensor_power_state(&hg_group->common_attributes, true);
> +		report_id = hg_state->hinge.report_id;
> +		min = hg_state->hinge.logical_minimum;
> +		if (report_id < 0) {
> +			*val = 0;
> +			hid_sensor_power_state(&hg_group->common_attributes,
> +					       false);
> +			return -EINVAL;
> +		}
> +
> +		*val = sensor_hub_input_attr_get_raw_value(
> +			hg_group->common_attributes.hsdev, hsdev->usage,
> +			hg_state->hinge_address, report_id, SENSOR_HUB_SYNC,
> +			min < 0);
> +
> +		hid_sensor_power_state(&hg_group->common_attributes, false);
> +		ret_type = IIO_VAL_INT;
> +		break;
> +	case IIO_CHAN_INFO_SCALE:
> +		*val = hg_state->scale_pre_decml;
> +		*val2 = hg_state->scale_post_decml;
> +		ret_type = hg_state->scale_precision;
> +		break;
> +	case IIO_CHAN_INFO_OFFSET:
> +		*val = hg_state->value_offset;
> +		ret_type = IIO_VAL_INT;
> +		break;
> +	case IIO_CHAN_INFO_SAMP_FREQ:
> +		ret_type = hid_sensor_read_samp_freq_value(
> +			&hg_group->common_attributes, val, val2);
> +		break;
> +	case IIO_CHAN_INFO_HYSTERESIS:
> +		ret_type = hid_sensor_read_raw_hyst_value(
> +			&hg_group->common_attributes, val, val2);
> +		break;
> +	default:
> +		ret_type = -EINVAL;
> +		break;
> +	}
> +
> +	return ret_type;
> +}
> +
> +/* Channel write_raw handler */
> +static int hinge_write_raw(struct iio_dev *indio_dev,
> +			   struct iio_chan_spec const *chan, int val, int val2,
> +			   long mask)
> +{
> +	int ret;
> +
> +	switch (mask) {
> +	case IIO_CHAN_INFO_SAMP_FREQ:
> +		ret = hid_sensor_write_samp_freq_value(
> +			&hg_group->common_attributes, val, val2);
> +		break;
> +	case IIO_CHAN_INFO_HYSTERESIS:
> +		ret = hid_sensor_write_raw_hyst_value(
> +			&hg_group->common_attributes, val, val2);
> +
> +		break;
> +	default:
> +		ret = -EINVAL;
> +	}
> +
> +	return ret;
> +}
> +
> +static const struct iio_info hinge_info = {
> +	.read_raw = &hinge_read_raw,
> +	.write_raw = &hinge_write_raw,
> +};
> +
> +/*
> + * Function to push data to buffer;
> + * wrapper added for symmetry with other hid-sensor drivers
> + */
> +static void hid_sensor_push_data(struct iio_dev *indio_dev, void *data, int len,

This doesn't seem to be generic, so don't name it as such.

> +				 int64_t timestamp)
> +{
> +	iio_push_to_buffers_with_timestamp(indio_dev, data, timestamp);
I hope that data buffer obeys the various rules needed by (and admittedly
not that well documented) iio_push_to_buffers_with_timestamp()

1. Needs to be 8 byte aligned.
2. Needs to have space for an aligned 8 byte timestamp at the end.

> +}
> +
> +/*
> + * Callback handler to send event after all samples are received
> + * and captured.
> + */
> +static int hinge_proc_event(struct hid_sensor_hub_device *hsdev,
> +			    unsigned int usage_id, void *priv)
> +{
> +	int i;
> +
> +	for (i = 0; i < IIO_DEV_NUM; ++i) {
If we push for all sensors together, better to have
this as a single iio_device with 3 channels.

Use the channel labels (just added to IIO) to identify which is which.

> +		struct hinge_state *hg_state;
> +		struct iio_dev *indio_dev;
> +
> +		hg_state = hg_group->hg_states[i];
> +		indio_dev = hg_state->indio_dev;
> +
> +		dev_dbg(&indio_dev->dev, "%s timestamp:%llu scan_bytes:%d\n",
> +			__func__, hg_state->timestamp, indio_dev->scan_bytes);
> +
> +		if (!hg_state->timestamp)
> +			hg_state->timestamp = iio_get_time_ns(indio_dev);
> +
> +		hid_sensor_push_data(indio_dev, hg_state->hinge_val,
> +				     sizeof(hg_state->hinge_val),
> +				     hg_state->timestamp);
> +
> +		hg_state->timestamp = 0;
> +	}
> +
> +	return 0;
> +}
> +
> +/* Capture samples in local storage */
> +static int hinge_capture_sample(struct hid_sensor_hub_device *hsdev,
> +				unsigned int usage_id, size_t raw_len,
> +				char *raw_data, void *priv)
> +{
> +	struct hinge_state *hg_state;
> +	int offset;
> +	int ret = -EINVAL;
> +	int i;
> +
> +	if (usage_id == HID_USAGE_SENSOR_TIME_TIMESTAMP) {
> +		for (i = 0; i < IIO_DEV_NUM; i++)
> +			hg_group->hg_states[i]->timestamp =

This rather implies all the data is captured together... If so single
iio_device may make more sense.

> +				hid_sensor_convert_timestamp(
> +					&hg_group->common_attributes,
> +					*(int64_t *)raw_data);
> +		return 0;
> +	}
> +
> +	switch (usage_id) {
> +	case HID_USAGE_SENSOR_DATA_FIELD_CUSTOM_VALUE_1:
> +	case HID_USAGE_SENSOR_DATA_FIELD_CUSTOM_VALUE_2:
> +	case HID_USAGE_SENSOR_DATA_FIELD_CUSTOM_VALUE_3:
> +		offset = usage_id - HID_USAGE_SENSOR_DATA_FIELD_CUSTOM_VALUE_1;
> +		hg_state = hg_group->hg_states[offset];
> +		hg_state->hinge_val[0] = *(u32 *)raw_data;
> +		ret = 0;

		return 0;

> +		break;
> +	default:
		return -EINVAL;
> +		break;
> +	}
> +
> +	return ret;
> +}
> +
> +/* Parse report which is specific to an usage id */
> +static int hinge_parse_report(struct platform_device *pdev,
> +			      struct hid_sensor_hub_device *hsdev,
> +			      unsigned int usage_id, unsigned int attr_usage_id,
> +			      struct hinge_state *st)
> +{
> +	int ret;
> +
> +	ret = sensor_hub_input_get_attribute_info(
> +		hsdev, HID_INPUT_REPORT, usage_id, attr_usage_id, &st->hinge);
> +	if (ret < 0)
> +		return ret;
> +
> +	st->hinge_address = attr_usage_id;
> +	st->scale_precision =
> +		hid_sensor_format_scale(HID_USAGE_SENSOR_HINGE, &st->hinge,
> +					&st->scale_pre_decml,
> +					&st->scale_post_decml);
> +
> +	/* Set Sensitivity field ids, when there is no individual modifier */
> +	if (hg_group->common_attributes.sensitivity.index < 0) {
> +		sensor_hub_input_get_attribute_info(
> +			hsdev, HID_FEATURE_REPORT, usage_id,
> +			HID_USAGE_SENSOR_DATA_MOD_CHANGE_SENSITIVITY_ABS |
> +				HID_USAGE_SENSOR_DATA_FIELD_CUSTOM_VALUE_1,
> +			&hg_group->common_attributes.sensitivity);
> +		dev_dbg(&pdev->dev, "Sensitivity index:report %d:%d\n",
> +			hg_group->common_attributes.sensitivity.index,
> +			hg_group->common_attributes.sensitivity.report_id);
> +	}
> +
> +	return ret;
> +}
> +
> +/* Function to initialize the processing for usage id */
> +static int hinge_add_iio_device(struct platform_device *pdev, int index,
> +				const char *name, struct hinge_state **st)
> +{
> +	struct hid_sensor_hub_device *hsdev = pdev->dev.platform_data;
> +	struct hinge_state *hg_state;
> +	struct iio_dev *indio_dev;
> +	int ret;
> +
> +	indio_dev =
> +		devm_iio_device_alloc(&pdev->dev, sizeof(struct hinge_state));

sizeof (*hg_state) preferred.

> +	if (indio_dev == NULL)
> +		return -ENOMEM;
> +
> +	hg_state = iio_priv(indio_dev);
> +	hg_state->indio_dev = indio_dev;
> +
> +	indio_dev->num_channels = ARRAY_SIZE(hinge_channels);
> +	indio_dev->channels =
> +		kmemdup(hinge_channels, sizeof(hinge_channels), GFP_KERNEL);

I don't immediately see anything that is modifying channels. As such you
should be able have it shared by all the instances.

> +	if (!indio_dev->channels)
> +		return -ENOMEM;
> +
> +	ret = hinge_parse_report(
> +		pdev, hsdev, hsdev->usage,
> +		HID_USAGE_SENSOR_DATA_FIELD_CUSTOM_VALUE_1 + index, hg_state);
> +	if (ret) {
> +		dev_err(&pdev->dev, "failed to setup attributes\n");
> +		goto error_free_dev_mem;
> +	}
> +
> +	indio_dev->dev.parent = &pdev->dev;
> +	indio_dev->info = &hinge_info;
> +	indio_dev->name = name;
> +	indio_dev->modes = INDIO_DIRECT_MODE;
> +
> +	ret = hid_sensor_setup_trigger(indio_dev, name,
> +				       &hg_group->common_attributes);
> +	if (ret < 0) {
> +		dev_err(&pdev->dev, "trigger setup failed\n");
> +		goto error_free_dev_mem;
> +	}
> +
> +	ret = iio_device_register(indio_dev);
> +	if (ret) {
> +		dev_err(&pdev->dev, "device register failed\n");
> +		goto error_remove_trigger;
> +	}
> +
> +	*st = hg_state;
> +
> +	return ret;
> +
> +error_remove_trigger:
> +	hid_sensor_remove_trigger(indio_dev, &hg_group->common_attributes);
> +error_free_dev_mem:
> +	kfree(indio_dev->channels);
> +	return ret;
> +}
> +
> +/* Function to deinitialize the processing for usage id */
> +static int hinge_remove_iio_device(struct platform_device *pdev, int index)
> +{
> +	struct hinge_state *hg_state = hg_group->hg_states[index];
> +	struct iio_dev *indio_dev = hg_state->indio_dev;
> +
> +	iio_device_unregister(indio_dev);
> +	hid_sensor_remove_trigger(indio_dev, &hg_group->common_attributes);
> +	kfree(indio_dev->channels);
> +
> +	return 0;
> +}
> +
> +static int hid_hinge_probe(struct platform_device *pdev)
> +{
> +	struct hinge_state *hg_state;
> +	struct hid_sensor_hub_device *hsdev = pdev->dev.platform_data;
> +	static const char *const names[] = { "hinge", "screen", "keyboard" };
> +	int ret;
> +	int i;
> +
> +	hg_group = devm_kzalloc(&pdev->dev, sizeof(struct hinge_group),
> +				GFP_KERNEL);

As mentioned above, I'd really not expect to see a global like this.
Technically nothing stops there being more than one instance of this
device on a platform (even if that would be a bit odd) + it's almost
always cleaner to not use a global in the first place.

> +	if (!hg_group)
> +		return -ENOMEM;
> +
> +	hg_group->common_attributes.hsdev = hsdev;
> +	hg_group->common_attributes.pdev = pdev;
> +
> +	ret = hid_sensor_parse_common_attributes(hsdev, hsdev->usage,
> +						 &hg_group->common_attributes);
> +	if (ret) {
> +		dev_err(&pdev->dev, "failed to setup common attributes\n");
> +		return ret;
> +	}
> +
> +	atomic_set(&hg_group->common_attributes.data_ready, 0);
> +	for (i = 0; i < IIO_DEV_NUM; i++) {
> +		ret = hinge_add_iio_device(pdev, i, names[i], &hg_state);
> +		if (ret)
> +			goto err_probe;
> +
> +		hg_group->hg_states[i] = hg_state;
> +	}
> +
> +	/* use the first iio device to do the PM */
> +	platform_set_drvdata(pdev, hg_group->hg_states[0]->indio_dev);
> +
> +	hg_group->callbacks.send_event = hinge_proc_event;
> +	hg_group->callbacks.capture_sample = hinge_capture_sample;
> +	hg_group->callbacks.pdev = pdev;
> +	ret = sensor_hub_register_callback(hsdev, hsdev->usage,
> +					   &hg_group->callbacks);
> +	if (ret < 0)
> +		dev_err(&pdev->dev, "callback reg failed\n");
> +
> +	return ret;
> +
> +err_probe:
> +	for (i--; i >= 0; i--)
> +		hinge_remove_iio_device(pdev, i);
> +
> +	return ret;
> +}
> +
> +/* Function to deinitialize the processing for usage id */
> +static int hid_hinge_remove(struct platform_device *pdev)
> +{
> +	struct hid_sensor_hub_device *hsdev = pdev->dev.platform_data;
> +	int i;
> +
> +	sensor_hub_remove_callback(hsdev, hsdev->usage);
> +
> +	for (i = 0; i < IIO_DEV_NUM; i++)
> +		hinge_remove_iio_device(pdev, i);
> +
> +	return 0;
> +}
> +
> +static const struct platform_device_id hid_hinge_ids[] = {
> +	{
> +		/* Format: HID-SENSOR-INT-usage_id_in_hex_lowercase */
> +		.name = "HID-SENSOR-INT-020b",
> +	},
> +	{ /* sentinel */ }
> +};
> +MODULE_DEVICE_TABLE(platform, hid_hinge_ids);
> +
> +static struct platform_driver hid_hinge_platform_driver = {
> +	.id_table = hid_hinge_ids,
> +	.driver = {
> +		.name	= KBUILD_MODNAME,
> +		.pm	= &hid_sensor_pm_ops,
> +	},
> +	.probe		= hid_hinge_probe,
> +	.remove		= hid_hinge_remove,
> +};
> +module_platform_driver(hid_hinge_platform_driver);
> +
> +MODULE_DESCRIPTION("HID Sensor Custom Hinge");
> +MODULE_AUTHOR("Ye Xiang <xiang.ye@intel.com>");
> +MODULE_LICENSE("GPL");