drivers: touchscreen: tsc2007 and ads7846/tsc2046 improvements (use common touchscreen bindings, pre-calibration, spi fix and provide iio raw values)

commit b98abe52fa8e ("Input: add common DT binding for touchscreens") introduced common DT bindings for touchscreens [1] and a helper function to parse the DT.

commit ed7c9870c9bc ("Input: of_touchscreen - add support for inverted / swapped axes") added another helper for parsing axis inversion and swapping
and applying them to x and y coordinates.

Both helpers have been integrated to accommodate any orientation of the touch panel in relation to the LCD.

A new feature is to introduce scaling the min/max ADC values to the screen size.

This makes it possible to pre-calibrate the touch so that is (almost) exactly matches the LCD pixel coordinates it is glued onto. This allows to well enough operate the touch before a user space calibration step can improve the precision.

Finally, calculate_pressure has been renamed to calculate_resistance because that is what it is doing.

[1]: Documentation/devicetree/bindings/input/touchscreen/touchscreen.txt

Signed-off-by H. Nikolaus Schaller <[email protected]>

---
 .../bindings/input/touchscreen/tsc2007.txt         |  20 ++--
 drivers/input/touchscreen/tsc2007.c                | 120 +++++++++++++++++----
 include/linux/i2c/tsc2007.h                        |   8 ++
 3 files changed, 118 insertions(+), 30 deletions(-)

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +1 -4

index ec365e1..6e9fd55 100644
@@ -6,6 +6,7 @@ Required properties:
 - ti,x-plate-ohms: X-plate resistance in ohms.
 
 Optional properties:
+- generic touch screen properties: see touchscreen binding [2].
 - gpios: the interrupt gpio the chip is connected to (trough the penirq pin).
   The penirq pin goes to low when the panel is touched.
   (see GPIO binding[1] for more details).

@@ -13,17 +14,20 @@ Optional properties:
   (see interrupt binding[0]).
 - interrupts: (gpio) interrupt to which the chip is connected
   (see interrupt binding[0]).
-- ti,max-rt: maximum pressure.
-- ti,fuzzx: specifies the absolute input fuzz x value.
-  If set, it will permit noise in the data up to +- the value given to the fuzz
-  parameter, that is used to filter noise from the event stream.

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +4 -1

 - ti,poll-period: how much time to wait (in milliseconds) before reading again the
-  values from the tsc2007.
+  values from the tsc2007 (default 1).
 
 [0]: Documentation/devicetree/bindings/interrupt-controller/interrupts.txt
 [1]: Documentation/devicetree/bindings/gpio/gpio.txt
+[2]: Documentation/devicetree/bindings/input/touchscreen/touchscreen.txt
 
 Example:
        &i2c1 {

@@ -35,6 +39,8 @@ Example:
                        interrupts = <0x0 0x8>;
                        gpios = <&gpio4 0 0>;
                        ti,x-plate-ohms = <180>;
+                       touchscreen-size-x = <640>;
+                       touchscreen-size-y = <480>;
                };
 
                /* ... */

Am 18.10.2016 um 18:22 schrieb Rob Herring <[email protected]>:

Quoted 4 lines · H. Nikolaus Schaller

commit b98abe52fa8e ("Input: add common DT binding for touchscreens") introduced common DT bindings for touchscreens [1] and a helper function to parse the DT.

Quoted 6 lines

commit ed7c9870c9bc ("Input: of_touchscreen - add support for inverted / swapped axes") added another helper for parsing axis inversion and swapping
and applying them to x and y coordinates.

Both helpers have been integrated to accommodate any orientation of the touch panel in relation to the LCD.

Please add the explanation of why this is a compatible change here.

Quoted 3 lines

A new feature is to introduce scaling the min/max ADC values to the screen size.

Sounds like a separate change.

Quoted 42 lines

This makes it possible to pre-calibrate the touch so that is (almost) exactly matches the LCD pixel coordinates it is glued onto. This allows to well enough operate the touch before a user space calibration step can improve the precision.

Finally, calculate_pressure has been renamed to calculate_resistance because that is what it is doing.

[1]: Documentation/devicetree/bindings/input/touchscreen/touchscreen.txt

Signed-off-by H. Nikolaus Schaller <[email protected]>

---

.../bindings/input/touchscreen/tsc2007.txt         |  20 ++--
drivers/input/touchscreen/tsc2007.c                | 120 +++++++++++++++++----
include/linux/i2c/tsc2007.h                        |   8 ++
3 files changed, 118 insertions(+), 30 deletions(-)

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +0 -1

index ec365e1..6e9fd55 100644
@@ -6,6 +6,7 @@ Required properties:
- ti,x-plate-ohms: X-plate resistance in ohms.
 

Optional properties:
+- generic touch screen properties: see touchscreen binding [2].
- gpios: the interrupt gpio the chip is connected to (trough the penirq pin). The penirq pin goes to low when the panel is touched.
(see GPIO binding[1] for more details).

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +2 -7

@@ -13,17 +14,20 @@ Optional properties:
  (see interrupt binding[0]).
- interrupts: (gpio) interrupt to which the chip is connected
  (see interrupt binding[0]).
-- ti,max-rt: maximum pressure.
-- ti,fuzzx: specifies the absolute input fuzz x value.
-  If set, it will permit noise in the data up to +- the value given to the fuzz
-  parameter, that is used to filter noise from the event stream.
-- ti,fuzzy: specifies the absolute input fuzz y value.
-- ti,fuzzz: specifies the absolute input fuzz z value.
+- ti,max-rt: maximum pressure resistance above which samples are ignored
+  (default: 4095).

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt Optional properties: · line 24

Quoted 6 lines

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +6 -0

+- ti,report-resistance: report resistance (no pressure = max_rt) instead
+  of pressure (no pressure = 0).
+- ti,min-x: minimum value reported by X axis ADC (default 0).
+- ti,max-x: maximum value reported by X axis ADC (default 4095).
+- ti,min-y: minimum value reported by Y axis ADC (default 0).
+- ti,max-y: maximum value reported by Y axis ADC (default 4095).

I thought these were going to be common? I think they should be.

Rob

Quoted 19 lines

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +1 -2

- ti,poll-period: how much time to wait (in milliseconds) before reading again the
-  values from the tsc2007.
+  values from the tsc2007 (default 1).
 

[0]: Documentation/devicetree/bindings/interrupt-controller/interrupts.txt [1]: Documentation/devicetree/bindings/gpio/gpio.txt
+[2]: Documentation/devicetree/bindings/input/touchscreen/touchscreen.txt

Example:

       &i2c1 {

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +2 -0

@@ -35,6 +39,8 @@ Example:
                       interrupts = <0x0 0x8>;
                       gpios = <&gpio4 0 0>;
                       ti,x-plate-ohms = <180>;
+                       touchscreen-size-x = <640>;
+                       touchscreen-size-y = <480>;
               };
 
               /* ... */

Hi Rob,

Quoted 11 lines

Am 18.10.2016 um 18:22 schrieb Rob Herring <[email protected]>:

Quoted 8 lines · H. Nikolaus Schaller

commit b98abe52fa8e ("Input: add common DT binding for touchscreens") introduced common DT bindings for touchscreens [1] and a helper function to parse the DT.

commit ed7c9870c9bc ("Input: of_touchscreen - add support for inverted / swapped axes") added another helper for parsing axis inversion and swapping
and applying them to x and y coordinates.

Quoted 4 lines

Quoted 2 lines

Both helpers have been integrated to accommodate any orientation of the touch panel in relation to the LCD.

Please add the explanation of why this is a compatible change here.

Can you please describe in more detail what you are missing here?

The patch simply makes use of the generic helper function introduced by the two patches cited above and is therefore automatically compatible if none of the new properties is defined.

The tsc2007 didn't have these features before. So what should we say about compatibility?

Quoted 15 lines

Quoted 4 lines

-- ti,max-rt: maximum pressure.
-- ti,fuzzx: specifies the absolute input fuzz x value.

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +0 -2

-  If set, it will permit noise in the data up to +- the value given to the fuzz
-  parameter, that is used to filter noise from the event stream.

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt Optional properties: · line 24

Quoted 10 lines

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +8 -2

-- ti,fuzzy: specifies the absolute input fuzz y value.
-- ti,fuzzz: specifies the absolute input fuzz z value.
+- ti,max-rt: maximum pressure resistance above which samples are ignored
+  (default: 4095).
+- ti,report-resistance: report resistance (no pressure = max_rt) instead
+  of pressure (no pressure = 0).
+- ti,min-x: minimum value reported by X axis ADC (default 0).
+- ti,max-x: maximum value reported by X axis ADC (default 4095).
+- ti,min-y: minimum value reported by Y axis ADC (default 0).
+- ti,max-y: maximum value reported by Y axis ADC (default 4095).

Quoted 1 line

I thought these were going to be common? I think they should be.

Yes, we had discussed that before.

In a second thought, I found that they are not even existing for the different touch chip drivers that are around. And some chip might not need them (because e.g. min/max are constants or defined my means outside the DT or the chip itself can store them in NV memory).

And, since they define ADC raw-values, they are very closely related to the individual chip, so that I am not sure if it is really necessary to make them common. At least if they are properly documented in the bindings for the specific chip they can have different names without disturbing each other.

So I think it is perfectly reasonable to define common bindings towards the input event layer (e.g. size, fuzz, swapping, inversion) and support them by common code which bakes raw coordinates into input events. The latest helper functions already do most of that fully automatic.

Next, I had simply copied them from the ads7846 driver where they exist for a long time. So this is sort of "common", at least for two different chips now. See also patch 6/8 of this series which adds the common properties to the ads7846 as well.

For more reference about the existing bindings:

Documentation/devicetree/bindings/input/ads7846.txt

(btw I think someone should move them to bindings/input/touchscreen).

It appears as if there is some overlap of the new generic properties (for x/y swapping) with the old ads7846 properties, but that is something you had already proposed a while ago to make me the driver recognize both properties to stay compatible with older DT files.

So if we now rename the min/max-x/y properties for the tsc2007 we have to rename them for the ads7846 (and maybe others) as well, which might break out-of-tree ads7846 devices or leads to more complex code for handling of property aliases.

Am 31.10.2016 um 04:39 schrieb Rob Herring <[email protected]>:

Quoted 16 lines · H. Nikolaus Schaller

Hi Rob,

Quoted 14 lines

Am 18.10.2016 um 18:22 schrieb Rob Herring <[email protected]>:

Quoted 10 lines · H. Nikolaus Schaller

commit b98abe52fa8e ("Input: add common DT binding for touchscreens") introduced common DT bindings for touchscreens [1] and a helper function to parse the DT.

commit ed7c9870c9bc ("Input: of_touchscreen - add support for inverted / swapped axes") added another helper for parsing axis inversion and swapping
and applying them to x and y coordinates.

Both helpers have been integrated to accommodate any orientation of the touch panel in relation to the LCD.

Quoted 3 lines

Quoted 1 line

Please add the explanation of why this is a compatible change here.

Can you please describe in more detail what you are missing here?

You stop handling ti,fuzzx for example. So the default driver value will be used and this is okay because...

Seems like you would have a regression in behavior if the fuzz values specified in an existing DT are ignored.

Quoted 7 lines

The patch simply makes use of the generic helper function introduced by the two patches cited above and is therefore automatically compatible if none of the new properties is defined.

The tsc2007 didn't have these features before. So what should we say about compatibility?

It certainly had the fuzz features before.

Quoted 6 lines

Quoted 6 lines

Quoted 6 lines

-- ti,max-rt: maximum pressure.
-- ti,fuzzx: specifies the absolute input fuzz x value.

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +0 -4

-  If set, it will permit noise in the data up to +- the value given to the fuzz
-  parameter, that is used to filter noise from the event stream.
-- ti,fuzzy: specifies the absolute input fuzz y value.
-- ti,fuzzz: specifies the absolute input fuzz z value.

Quoted 18 lines

Quoted 10 lines

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt Optional properties: · line 24

Quoted 8 lines

+- ti,max-rt: maximum pressure resistance above which samples are ignored

Documentation/devicetree/bindings/input/touchscreen/tsc2007.txt +7 -0

+  (default: 4095).
+- ti,report-resistance: report resistance (no pressure = max_rt) instead
+  of pressure (no pressure = 0).
+- ti,min-x: minimum value reported by X axis ADC (default 0).
+- ti,max-x: maximum value reported by X axis ADC (default 4095).
+- ti,min-y: minimum value reported by Y axis ADC (default 0).
+- ti,max-y: maximum value reported by Y axis ADC (default 4095).

I thought these were going to be common? I think they should be.

Yes, we had discussed that before.

In a second thought, I found that they are not even existing for the different touch chip drivers that are around. And some chip might not need them (because e.g. min/max are constants or defined my means outside the DT or the chip itself can store them in NV memory).

Quoted 5 lines

And, since they define ADC raw-values, they are very closely related to the individual chip, so that I am not sure if it is really necessary to make them common. At least if they are properly documented in the bindings for the specific chip they can have different names without disturbing each other.

They wouldn't disturb each other, but then there is no chance to have common parsing either.

Quoted 5 lines

So I think it is perfectly reasonable to define common bindings towards the input event layer (e.g. size, fuzz, swapping, inversion) and support them by common code which bakes raw coordinates into input events. The latest helper functions already do most of that fully automatic.

I could argue that size in pixels has no business being in DT.

That's a property of the panel the touchscreen is glued to. I guess in some cases, the controller is internally scaling the ADC values.

Quoted 6 lines

Next, I had simply copied them from the ads7846 driver where they exist for a long time. So this is sort of "common", at least for two different chips now. See also patch 6/8 of this series which adds the common properties to the ads7846 as well.

Quoted 5 lines

For more reference about the existing bindings:

Documentation/devicetree/bindings/input/ads7846.txt

(btw I think someone should move them to bindings/input/touchscreen).

Patches welcome.

Quoted 9 lines

It appears as if there is some overlap of the new generic properties (for x/y swapping) with the old ads7846 properties, but that is something you had already proposed a while ago to make me the driver recognize both properties to stay compatible with older DT files.

So if we now rename the min/max-x/y properties for the tsc2007 we have to rename them for the ads7846 (and maybe others) as well, which might break out-of-tree ads7846 devices or leads to more complex code for handling of property aliases.

We don't have to rename them for ads7846. That's already baked.

But there's a trend here.

You're defining the same property. Use a common
name, so when the 3rd binding comes along needing the same thing, we already have a common binding.

The tsc2007 chip not only has a resistive touch screen controller but also an external AUX adc imput which can be used for an ambient light sensor, battery voltage monitoring or any general purpose.

Additionally it can measure the chip temperature.

This extension provides an iio interface for these adc channels.

Since it is not wasting much resources and is very straightforward, we simply provide all other adc channels as optional iio interfaces as weel. This can be used for debugging or special applications.

Signed-off-by H. Nikolaus Schaller <[email protected]>

---
 drivers/input/touchscreen/tsc2007.c | 151 +++++++++++++++++++++++++++++++++++-
 1 file changed, 150 insertions(+), 1 deletion(-)

drivers/input/touchscreen/tsc2007.c +150 -1

index c314331..b4a9504 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>
 
 #define TSC2007_MEASURE_TEMP0		(0x0 << 4)
 #define TSC2007_MEASURE_AUX		(0x2 << 4)

@@ -61,6 +62,16 @@
 #define READ_X		(ADC_ON_12BIT | TSC2007_MEASURE_X)
 #define PWRDOWN		(TSC2007_12BIT | TSC2007_POWER_OFF_IRQ_EN)
 
+#define TSC2007_CHAN_IIO(_chan, _name, _type, _chan_info) \
+{ \
+	.datasheet_name = _name, \
+	.type = _type, \
+	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |	\
+			BIT(_chan_info), \
+	.indexed = 1, \
+	.channel = _chan, \
+}
+
 struct ts_event {
 	u16	x;
 	u16	y;

@@ -69,9 +80,13 @@ struct ts_event {
 
 struct tsc2007 {
 	struct input_dev	*input;
+#ifdef CONFIG_IIO
+	struct iio_dev		*indio;
+#endif
 	char			phys[32];
 
 	struct i2c_client	*client;
+	struct mutex		mlock;
 
 	u16			model;
 	u16			x_plate_ohms;

@@ -192,7 +207,10 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
 	while (!ts->stopped && tsc2007_is_pen_down(ts)) {
 
 		/* pen is down, continue with the measurement */
+
+		mutex_lock(&ts->mlock);
 		tsc2007_read_values(ts, &tc);
+		mutex_unlock(&ts->mlock);
 
 		rt = tsc2007_calculate_resistance(ts, &tc);
 

@@ -319,6 +337,89 @@ static void tsc2007_close(struct input_dev *input_dev)
 	tsc2007_stop(ts);
 }
 
+#ifdef CONFIG_IIO
+
+static const struct iio_chan_spec tsc2007_iio_channel[] = {
+	TSC2007_CHAN_IIO(0, "x", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(1, "y", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(2, "z1", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(3, "z2", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(4, "adc", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(5, "rt", IIO_VOLTAGE, IIO_CHAN_INFO_RAW), /* Ohms? */
+	TSC2007_CHAN_IIO(6, "pen", IIO_PRESSURE, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(7, "temp0", IIO_TEMP, IIO_CHAN_INFO_RAW),
+	TSC2007_CHAN_IIO(8, "temp1", IIO_TEMP, IIO_CHAN_INFO_RAW),
+};
+
+static int tsc2007_read_raw(struct iio_dev *indio_dev,
+	struct iio_chan_spec const *chan, int *val, int *val2, long mask)
+{
+	struct  tsc2007 *tsc = iio_priv(indio_dev);
+	int adc_chan = chan->channel;
+	int ret = 0;
+
+	if (adc_chan >= ARRAY_SIZE(tsc2007_iio_channel))
+		return -EINVAL;
+
+	if (mask != IIO_CHAN_INFO_RAW)
+		return -EINVAL;
+
+	mutex_lock(&tsc->mlock);
+
+	switch (chan->channel) {
+	case 0:
+		*val = tsc2007_xfer(tsc, READ_X);
+		break;
+	case 1:
+		*val = tsc2007_xfer(tsc, READ_Y);
+		break;
+	case 2:
+		*val = tsc2007_xfer(tsc, READ_Z1);
+		break;
+	case 3:
+		*val = tsc2007_xfer(tsc, READ_Z2);
+		break;
+	case 4:
+		*val = tsc2007_xfer(tsc, (ADC_ON_12BIT | TSC2007_MEASURE_AUX));
+		break;
+	case 5: {
+		struct ts_event tc;
+
+		tc.x = tsc2007_xfer(tsc, READ_X);
+		tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+		tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+		*val = tsc2007_calculate_resistance(tsc, &tc);
+		break;
+	}
+	case 6:
+		*val = tsc2007_is_pen_down(tsc);
+		break;
+	case 7:
+		*val = tsc2007_xfer(tsc,
+				    (ADC_ON_12BIT | TSC2007_MEASURE_TEMP0));
+		break;
+	case 8:
+		*val = tsc2007_xfer(tsc,
+				    (ADC_ON_12BIT | TSC2007_MEASURE_TEMP1));
+		break;
+	}
+
+	/* Prepare for next touch reading - power down ADC, enable PENIRQ */
+	tsc2007_xfer(tsc, PWRDOWN);
+
+	mutex_unlock(&tsc->mlock);
+
+	ret = IIO_VAL_INT;
+
+	return ret;
+}
+
+static const struct iio_info tsc2007_iio_info = {
+	.read_raw = tsc2007_read_raw,
+	.driver_module = THIS_MODULE,
+};
+#endif
+
 #ifdef CONFIG_OF
 static int tsc2007_get_pendown_state_gpio(struct device *dev)
 {

@@ -453,26 +554,56 @@ static int tsc2007_probe(struct i2c_client *client,
 	const struct tsc2007_platform_data *pdata = dev_get_platdata(&client->dev);
 	struct tsc2007 *ts;
 	struct input_dev *input_dev;
+#ifdef CONFIG_IIO
+	struct iio_dev *indio_dev;
+#endif
 	int err;
 
 	if (!i2c_check_functionality(client->adapter,
 				     I2C_FUNC_SMBUS_READ_WORD_DATA))
 		return -EIO;
 
+#ifdef CONFIG_IIO
+	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));
+	if (!indio_dev) {
+		dev_err(&client->dev, "iio_device_alloc failed\n");
+		return -ENOMEM;
+	}
+
+	ts = iio_priv(indio_dev);
+#else
 	ts = devm_kzalloc(&client->dev, sizeof(struct tsc2007), GFP_KERNEL);
 	if (!ts)
 		return -ENOMEM;
-
+#endif
 	input_dev = devm_input_allocate_device(&client->dev);
 	if (!input_dev)
 		return -ENOMEM;
 
 	i2c_set_clientdata(client, ts);
 
+#ifdef CONFIG_IIO
+	indio_dev->name = "tsc2007";
+	indio_dev->dev.parent = &client->dev;
+	indio_dev->info = &tsc2007_iio_info;
+	indio_dev->modes = INDIO_DIRECT_MODE;
+	indio_dev->channels = tsc2007_iio_channel;
+	indio_dev->num_channels = ARRAY_SIZE(tsc2007_iio_channel);
+
+	err = iio_device_register(indio_dev);
+	if (err < 0) {
+		dev_err(&client->dev, "iio_device_register() failed: %d\n",
+			err);
+		return err;
+	}
+	ts->indio = indio_dev;
+#endif
+
 	ts->client = client;
 	ts->irq = client->irq;
 	ts->input = input_dev;
 	init_waitqueue_head(&ts->wait);
+	mutex_init(&ts->mlock);
 
 	snprintf(ts->phys, sizeof(ts->phys),
 		 "%s/input0", dev_name(&client->dev));

@@ -543,7 +674,9 @@ static int tsc2007_probe(struct i2c_client *client,
 	if (err < 0) {
 		dev_err(&client->dev,
 			"Failed to setup chip: %d\n", err);
+#ifdef CONFIG_IIO
 		iio_device_unregister(indio_dev);
+#endif
 		return err;	/* usually, chip does not respond */
 	}
 

@@ -557,6 +690,21 @@ static int tsc2007_probe(struct i2c_client *client,
 	return 0;
 }
 
+static int tsc2007_remove(struct i2c_client *client)
+{
+	struct tsc2007 *ts = i2c_get_clientdata(client);
+	struct input_dev *input_dev = ts->input;
+#ifdef CONFIG_IIO
+	struct iio_dev *indio_dev = ts->indio;
+#endif
+	input_unregister_device(input_dev);
+
+#ifdef CONFIG_IIO
+	iio_device_unregister(indio_dev);
+#endif
+	return 0;
+}
+
 static const struct i2c_device_id tsc2007_idtable[] = {
 	{ "tsc2007", 0 },
 	{ }

@@ -579,6 +727,7 @@ static struct i2c_driver tsc2007_driver = {
 	},
 	.id_table	= tsc2007_idtable,
 	.probe		= tsc2007_probe,
+	.remove		= tsc2007_remove,
 };
 
 module_i2c_driver(tsc2007_driver);

Am 22.10.2016 um 20:33 schrieb Jonathan Cameron <[email protected]>:

Quoted 11 lines · H. Nikolaus Schaller

The tsc2007 chip not only has a resistive touch screen controller but also an external AUX adc imput which can be used for an ambient light sensor, battery voltage monitoring or any general purpose.

Additionally it can measure the chip temperature.

This extension provides an iio interface for these adc channels.

Since it is not wasting much resources and is very straightforward, we simply provide all other adc channels as optional iio interfaces as weel. This can be used for debugging or special applications.

well

Quoted 2 lines

Signed-off-by H. Nikolaus Schaller <[email protected]>

This could be cleaner done perhaps by factoring out the IIO stuff into a separate file and using a header with stubs to deal with the no available case.

There will only be a handful of stubs and it'll give you a lot cleaner code in here.

If def fun in .c files is always harder to deal with than in a header where stubs are really obvious.

Hi Jonathan,

Quoted 17 lines

Am 22.10.2016 um 20:33 schrieb Jonathan Cameron <[email protected]>:

Quoted 11 lines · H. Nikolaus Schaller

The tsc2007 chip not only has a resistive touch screen controller but also an external AUX adc imput which can be used for an ambient light sensor, battery voltage monitoring or any general purpose.

Additionally it can measure the chip temperature.

This extension provides an iio interface for these adc channels.

Since it is not wasting much resources and is very straightforward, we simply provide all other adc channels as optional iio interfaces as weel. This can be used for debugging or special applications.

well

Quoted 2 lines

Signed-off-by H. Nikolaus Schaller <[email protected]>

Quoted 8 lines

This could be cleaner done perhaps by factoring out the IIO stuff into a separate file and using a header with stubs to deal with the no available case.

There will only be a handful of stubs and it'll give you a lot cleaner code in here.

If def fun in .c files is always harder to deal with than in a header where stubs are really obvious.

Yes, it became a lot of #ifdefs spread over the source file.

The easiest thing would be to require IIO to be enabled :)

With your proposal to consider refactoring, I think the crucial part is the conditional allocation either through devm_iio_device_alloc() or devm_kzalloc(). This can be refactored into some conditional tsc2007_alloc().

I have tried some draft (not tested and not tidied up) to check if the direction is good.

This reduces the number of #ifdef CONFIG_IIO from 7 to 2 without introducing new files or includes. There are also 2 other #ifdef CONFIG_OF so it doesn't seem to be very complex now in comparison. And the patch itself has only a handful of hunks (8).

Moving tsc2007_alloc into a separate file tsc2007_iio.c would only move around one #ifdef CONFIG_OF from tsc2007.c but IMHO makes it more difficult to understand because it is not really iio specific and one has to switch between two source files. And I would have to touch the Makefile as well.

What do you think?

If generally ok, I can include that in [PATCH v5].

BR and thanks,
Nikolaus

drivers/input/touchscreen/tsc2007.c +1 -0

index 5e3c4bf..691e79f 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>
 
 #define TSC2007_MEASURE_TEMP0          (0x0 << 4)
 #define TSC2007_MEASURE_AUX            (0x2 << 4)

drivers/input/touchscreen/tsc2007.c +93 -0

        char                    phys[32];
 
        struct i2c_client       *client;
+       struct mutex            mlock;
 
        u16                     model;
        u16                     x_plate_ohms;

@@ -192,7 +197,10 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
        while (!ts->stopped && tsc2007_is_pen_down(ts)) {
 
                /* pen is down, continue with the measurement */
+
+               mutex_lock(&ts->mlock);
                tsc2007_read_values(ts, &tc);
+               mutex_unlock(&ts->mlock);
 
                rt = tsc2007_calculate_resistance(ts, &tc);
 

@@ -319,6 +327,162 @@ static void tsc2007_close(struct input_dev *input_dev)
        tsc2007_stop(ts);
 }
 
+#ifdef CONFIG_IIO
+
+#define TSC2007_CHAN_IIO(_chan, _name, _type, _chan_info) \
+{ \
+       .datasheet_name = _name, \
+       .type = _type, \
+       .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |  \
+                       BIT(_chan_info), \
+       .indexed = 1, \
+       .channel = _chan, \
+}
+
+static const struct iio_chan_spec tsc2007_iio_channel[] = {
+       TSC2007_CHAN_IIO(0, "x", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(1, "y", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(2, "z1", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(3, "z2", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(4, "adc", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(5, "rt", IIO_VOLTAGE, IIO_CHAN_INFO_RAW), /* Ohms? */
+       TSC2007_CHAN_IIO(6, "pen", IIO_PRESSURE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(7, "temp0", IIO_TEMP, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(8, "temp1", IIO_TEMP, IIO_CHAN_INFO_RAW),
+};
+
+static int tsc2007_read_raw(struct iio_dev *indio_dev,
+       struct iio_chan_spec const *chan, int *val, int *val2, long mask)
+{
+       struct  tsc2007 *tsc = iio_priv(indio_dev);
+       int adc_chan = chan->channel;
+       int ret = 0;
+
+       if (adc_chan >= ARRAY_SIZE(tsc2007_iio_channel))
+               return -EINVAL;
+
+       if (mask != IIO_CHAN_INFO_RAW)
+               return -EINVAL;
+
+       mutex_lock(&tsc->mlock);
+
+       switch (chan->channel) {
+       case 0:
+               *val = tsc2007_xfer(tsc, READ_X);
+               break;
+       case 1:
+               *val = tsc2007_xfer(tsc, READ_Y);
+               break;
+       case 2:
+               *val = tsc2007_xfer(tsc, READ_Z1);
+               break;
+       case 3:
+               *val = tsc2007_xfer(tsc, READ_Z2);
+               break;
+       case 4:
+               *val = tsc2007_xfer(tsc, (ADC_ON_12BIT | TSC2007_MEASURE_AUX));
+               break;
+       case 5: {
+               struct ts_event tc;
+
+               tc.x = tsc2007_xfer(tsc, READ_X);
+               tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+               tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+               *val = tsc2007_calculate_resistance(tsc, &tc);
+               break;
+       }
+       case 6:
+               *val = tsc2007_is_pen_down(tsc);
+               break;
+       case 7:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP0));
+               break;
+       case 8:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP1));
+               break;
+       }
+
+       /* Prepare for next touch reading - power down ADC, enable PENIRQ */
+       tsc2007_xfer(tsc, PWRDOWN);
+
+       mutex_unlock(&tsc->mlock);
+
+       ret = IIO_VAL_INT;
+
+       return ret;
+}
+
+static const struct iio_info tsc2007_iio_info = {
+       .read_raw = tsc2007_read_raw,

drivers/input/touchscreen/tsc2007.c +44 -0

+       if (!indio_dev) {
+               dev_err(&client->dev, "iio_device_alloc failed\n");
+               return -ENOMEM;
+       }
+
+       *ts = iio_priv(indio_dev);
+
+       *input_dev = devm_input_allocate_device(&client->dev);
+       if (!*input_dev)
+               return -ENOMEM;
+
+       i2c_set_clientdata(client, *ts);
+       (*ts)->indio = indio_dev;
+
+       indio_dev->name = "tsc2007";
+       indio_dev->dev.parent = &client->dev;
+       indio_dev->info = &tsc2007_iio_info;
+       indio_dev->modes = INDIO_DIRECT_MODE;
+       indio_dev->channels = tsc2007_iio_channel;
+       indio_dev->num_channels = ARRAY_SIZE(tsc2007_iio_channel);
+
+       err = iio_device_register(indio_dev);
+       if (err < 0) {
+               dev_err(&client->dev, "iio_device_register() failed: %d\n",
+                       err);
+               return err;
+       }
+
+       return 0;
+}
+
+#define tsc2007_iio_device_unregister(ts) iio_device_unregister(ts->indio)
+
+#else /* CONFIG_IIO */
+
+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+
+       *ts = devm_kzalloc(&client->dev, sizeof(struct tsc2007), GFP_KERNEL);
+       if (!*ts)
+               return -ENOMEM;
+

drivers/input/touchscreen/tsc2007.c +17 -9

+
+#endif /* CONFIG_IIO */
+
 #ifdef CONFIG_OF
 static int tsc2007_get_pendown_state_gpio(struct device *dev)
 {

@@ -459,20 +623,15 @@ static int tsc2007_probe(struct i2c_client *client,
                                     I2C_FUNC_SMBUS_READ_WORD_DATA))
                return -EIO;
 
-       ts = devm_kzalloc(&client->dev, sizeof(struct tsc2007), GFP_KERNEL);
-       if (!ts)
-               return -ENOMEM;
-
-       input_dev = devm_input_allocate_device(&client->dev);
-       if (!input_dev)
-               return -ENOMEM;
-
-       i2c_set_clientdata(client, ts);
+       err = tsc2007_alloc(client, &ts, &input_dev);
+       if (err < 0)
+               return err;
 
        ts->client = client;
        ts->irq = client->irq;
        ts->input = input_dev;
        init_waitqueue_head(&ts->wait);
+       mutex_init(&ts->mlock);
 
        snprintf(ts->phys, sizeof(ts->phys),
                 "%s/input0", dev_name(&client->dev));

@@ -543,6 +702,7 @@ static int tsc2007_probe(struct i2c_client *client,
        if (err < 0) {
                dev_err(&client->dev,
                        "Failed to setup chip: %d\n", err);
+               tsc2007_iio_device_unregister(ts);
                return err;     /* usually, chip does not respond */
        }
 

@@ -556,6 +716,14 @@ static int tsc2007_probe(struct i2c_client *client,
        return 0;
 }
 
+static int tsc2007_remove(struct i2c_client *client)
+{
+       struct tsc2007 *ts = i2c_get_clientdata(client);
+       input_unregister_device(ts->input);
+       tsc2007_iio_device_unregister(ts);
+       return 0;
+}
+
 static const struct i2c_device_id tsc2007_idtable[] = {
        { "tsc2007", 0 },
        { }

@@ -578,6 +746,7 @@ static struct i2c_driver tsc2007_driver = {
        },
        .id_table       = tsc2007_idtable,
        .probe          = tsc2007_probe,
+       .remove         = tsc2007_remove,
 };
 
 module_i2c_driver(tsc2007_driver);

Am 23.10.2016 um 11:24 schrieb Jonathan Cameron <[email protected]>:

Quoted 50 lines · H. Nikolaus Schaller

Hi Jonathan,

Quoted 17 lines

Am 22.10.2016 um 20:33 schrieb Jonathan Cameron <[email protected]>:

Quoted 11 lines · H. Nikolaus Schaller

The tsc2007 chip not only has a resistive touch screen controller but also an external AUX adc imput which can be used for an ambient light sensor, battery voltage monitoring or any general purpose.

Additionally it can measure the chip temperature.

This extension provides an iio interface for these adc channels.

Since it is not wasting much resources and is very straightforward, we simply provide all other adc channels as optional iio interfaces as weel. This can be used for debugging or special applications.

well

Quoted 2 lines

Signed-off-by H. Nikolaus Schaller <[email protected]>

Quoted 8 lines

This could be cleaner done perhaps by factoring out the IIO stuff into a separate file and using a header with stubs to deal with the no available case.

There will only be a handful of stubs and it'll give you a lot cleaner code in here.

If def fun in .c files is always harder to deal with than in a header where stubs are really obvious.

Yes, it became a lot of #ifdefs spread over the source file.

The easiest thing would be to require IIO to be enabled :)

With your proposal to consider refactoring, I think the crucial part is the conditional allocation either through devm_iio_device_alloc() or devm_kzalloc(). This can be refactored into some conditional tsc2007_alloc().

I have tried some draft (not tested and not tidied up) to check if the direction is good.

This reduces the number of #ifdef CONFIG_IIO from 7 to 2 without introducing new files or includes. There are also 2 other #ifdef CONFIG_OF so it doesn't seem to be very complex now in comparison. And the patch itself has only a handful of hunks (8).

Moving tsc2007_alloc into a separate file tsc2007_iio.c would only move around one #ifdef CONFIG_OF from tsc2007.c but IMHO makes it more difficult to understand because it is not really iio specific and one has to switch between two source files. And I would have to touch the Makefile as well.

Quoted 1 line

What do you think?

I'd still split it. The only bit of the IIO block that isn't specific is a tiny chunk of the allocation code (as you've highlighted).

Even that can be avoided by adding a tiny bit more indirection than would otherwise be needed (it's not pretty but it would give a clean separation).

It's pretty much the way this sort of optional functionality should always be done - means that if you don't care (i.e. it's not enabled) you don't even have to see the code.

Jonathan

Quoted 21 lines

If generally ok, I can include that in [PATCH v5].

BR and thanks,
Nikolaus

drivers/input/touchscreen/tsc2007.c +1 -0

index 5e3c4bf..691e79f 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>
 
 #define TSC2007_MEASURE_TEMP0          (0x0 << 4)
 #define TSC2007_MEASURE_AUX            (0x2 << 4)

@@ -69,9 +70,13 @@ struct ts_event {
 

drivers/input/touchscreen/tsc2007.c struct ts_event { · line 75

Quoted 5 lines

drivers/input/touchscreen/tsc2007.c +3 -0

 struct tsc2007 {
       struct input_dev        *input;
+#ifdef CONFIG_IIO
+       struct iio_dev          *indio;
+#endif

I wouldn't bother with this one. Just have struct iio_dev; before this and it'll waste a whole one pointer (+ you shouldn't need to have iio.h included in here once you have spit the files).

Quoted 111 lines

drivers/input/touchscreen/tsc2007.c +93 -0

       char                    phys[32];
 
       struct i2c_client       *client;
+       struct mutex            mlock;
 
       u16                     model;
       u16                     x_plate_ohms;

@@ -192,7 +197,10 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
       while (!ts->stopped && tsc2007_is_pen_down(ts)) {
 
               /* pen is down, continue with the measurement */
+
+               mutex_lock(&ts->mlock);
               tsc2007_read_values(ts, &tc);
+               mutex_unlock(&ts->mlock);
 
               rt = tsc2007_calculate_resistance(ts, &tc);
 

@@ -319,6 +327,162 @@ static void tsc2007_close(struct input_dev *input_dev)
       tsc2007_stop(ts);
 }
 
+#ifdef CONFIG_IIO
+
+#define TSC2007_CHAN_IIO(_chan, _name, _type, _chan_info) \
+{ \
+       .datasheet_name = _name, \
+       .type = _type, \
+       .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |  \
+                       BIT(_chan_info), \
+       .indexed = 1, \
+       .channel = _chan, \
+}
+
+static const struct iio_chan_spec tsc2007_iio_channel[] = {
+       TSC2007_CHAN_IIO(0, "x", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(1, "y", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(2, "z1", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(3, "z2", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(4, "adc", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(5, "rt", IIO_VOLTAGE, IIO_CHAN_INFO_RAW), /* Ohms? */
+       TSC2007_CHAN_IIO(6, "pen", IIO_PRESSURE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(7, "temp0", IIO_TEMP, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(8, "temp1", IIO_TEMP, IIO_CHAN_INFO_RAW),
+};
+
+static int tsc2007_read_raw(struct iio_dev *indio_dev,
+       struct iio_chan_spec const *chan, int *val, int *val2, long mask)
+{
+       struct  tsc2007 *tsc = iio_priv(indio_dev);
+       int adc_chan = chan->channel;
+       int ret = 0;
+
+       if (adc_chan >= ARRAY_SIZE(tsc2007_iio_channel))
+               return -EINVAL;
+
+       if (mask != IIO_CHAN_INFO_RAW)
+               return -EINVAL;
+
+       mutex_lock(&tsc->mlock);
+
+       switch (chan->channel) {
+       case 0:
+               *val = tsc2007_xfer(tsc, READ_X);
+               break;
+       case 1:
+               *val = tsc2007_xfer(tsc, READ_Y);
+               break;
+       case 2:
+               *val = tsc2007_xfer(tsc, READ_Z1);
+               break;
+       case 3:
+               *val = tsc2007_xfer(tsc, READ_Z2);
+               break;
+       case 4:
+               *val = tsc2007_xfer(tsc, (ADC_ON_12BIT | TSC2007_MEASURE_AUX));
+               break;
+       case 5: {
+               struct ts_event tc;
+
+               tc.x = tsc2007_xfer(tsc, READ_X);
+               tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+               tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+               *val = tsc2007_calculate_resistance(tsc, &tc);
+               break;
+       }
+       case 6:
+               *val = tsc2007_is_pen_down(tsc);
+               break;
+       case 7:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP0));
+               break;
+       case 8:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP1));
+               break;
+       }
+
+       /* Prepare for next touch reading - power down ADC, enable PENIRQ */
+       tsc2007_xfer(tsc, PWRDOWN);
+
+       mutex_unlock(&tsc->mlock);
+
+       ret = IIO_VAL_INT;
+
+       return ret;
+}
+
+static const struct iio_info tsc2007_iio_info = {
+       .read_raw = tsc2007_read_raw,

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev) · line 428

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c +10 -0

+       .driver_module = THIS_MODULE,
+};
+
+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev) · line 430

Quoted 2 lines

drivers/input/touchscreen/tsc2007.c +1 -0

+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));
 	*ts = iio_priv(indio_dev);

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev) · line 431

Quoted 1 line

drivers/input/touchscreen/tsc2007.c +0 -0

 	 *ts = devm_kzalloc(&client->dev, sizeof(struct tsc2007), GFP_KERNEL);

Having doing that, you can have this CONFIG_IIO block as just doing the iio stuff with the input elements pulled back into the main probe function.

Then define something like

iio_configure (stubbed to nothing if no IIO) and
iio_unconfigure (also stubbed to nothing if no IIO).

A couple of additions in the header to make it all work (the struct tsc2007 and tsc2007_xfer() + a few of the register defines..

Nothing big and gets all the CONFIG_IIO into some really obvious stubbing out in the header.

Quoted 49 lines

drivers/input/touchscreen/tsc2007.c +48 -0

+       if (!indio_dev) {
+               dev_err(&client->dev, "iio_device_alloc failed\n");
+               return -ENOMEM;
+       }
+
+       *ts = iio_priv(indio_dev);
+
+       *input_dev = devm_input_allocate_device(&client->dev);
+       if (!*input_dev)
+               return -ENOMEM;
+
+       i2c_set_clientdata(client, *ts);
+       (*ts)->indio = indio_dev;
+
+       indio_dev->name = "tsc2007";
+       indio_dev->dev.parent = &client->dev;
+       indio_dev->info = &tsc2007_iio_info;
+       indio_dev->modes = INDIO_DIRECT_MODE;
+       indio_dev->channels = tsc2007_iio_channel;
+       indio_dev->num_channels = ARRAY_SIZE(tsc2007_iio_channel);
+
+       err = iio_device_register(indio_dev);
+       if (err < 0) {
+               dev_err(&client->dev, "iio_device_register() failed: %d\n",
+                       err);
+               return err;
+       }
+
+       return 0;
+}
+
+#define tsc2007_iio_device_unregister(ts) iio_device_unregister(ts->indio)
+
+#else /* CONFIG_IIO */
+
+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+
+       *ts = devm_kzalloc(&client->dev, sizeof(struct tsc2007), GFP_KERNEL);
+       if (!*ts)
+               return -ENOMEM;
+
+       *input_dev = devm_input_allocate_device(&client->dev);
+       if (!*input_dev)
+               return -ENOMEM;
+

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev) · line 482

Quoted 6 lines

drivers/input/touchscreen/tsc2007.c +6 -0

+       i2c_set_clientdata(client, *ts);
+
+       return 0;
+}
+
+#define tsc2007_iio_device_unregister(ts) /* not needed */

That's rather ugly and fragile. I'd stub it out as an actual function with no content and let the compiler drop it.

Quoted 61 lines

drivers/input/touchscreen/tsc2007.c +3 -0

+
+#endif /* CONFIG_IIO */
+

#ifdef CONFIG_OF
static int tsc2007_get_pendown_state_gpio(struct device *dev) {

drivers/input/touchscreen/tsc2007.c +14 -9

@@ -459,20 +623,15 @@ static int tsc2007_probe(struct i2c_client *client,
                                    I2C_FUNC_SMBUS_READ_WORD_DATA))
               return -EIO;
 
-       ts = devm_kzalloc(&client->dev, sizeof(struct tsc2007), GFP_KERNEL);
-       if (!ts)
-               return -ENOMEM;
-
-       input_dev = devm_input_allocate_device(&client->dev);
-       if (!input_dev)
-               return -ENOMEM;
-
-       i2c_set_clientdata(client, ts);
+       err = tsc2007_alloc(client, &ts, &input_dev);
+       if (err < 0)
+               return err;
 
       ts->client = client;
       ts->irq = client->irq;
       ts->input = input_dev;
       init_waitqueue_head(&ts->wait);
+       mutex_init(&ts->mlock);
 
       snprintf(ts->phys, sizeof(ts->phys),
                "%s/input0", dev_name(&client->dev));

@@ -543,6 +702,7 @@ static int tsc2007_probe(struct i2c_client *client,
       if (err < 0) {
               dev_err(&client->dev,
                       "Failed to setup chip: %d\n", err);
+               tsc2007_iio_device_unregister(ts);
               return err;     /* usually, chip does not respond */
       }
 

@@ -556,6 +716,14 @@ static int tsc2007_probe(struct i2c_client *client,
       return 0;
 }
 
+static int tsc2007_remove(struct i2c_client *client)
+{
+       struct tsc2007 *ts = i2c_get_clientdata(client);
+       input_unregister_device(ts->input);
+       tsc2007_iio_device_unregister(ts);
+       return 0;
+}
+
 static const struct i2c_device_id tsc2007_idtable[] = {
       { "tsc2007", 0 },
       { }

@@ -578,6 +746,7 @@ static struct i2c_driver tsc2007_driver = {
       },
       .id_table       = tsc2007_idtable,
       .probe          = tsc2007_probe,
+       .remove         = tsc2007_remove,
 };
 

drivers/input/touchscreen/tsc2007.c static struct i2c_driver tsc2007_driver = { · line 752

Quoted 2 lines

drivers/input/touchscreen/tsc2007.c +0 -0

 module_i2c_driver(tsc2007_driver);

Am 23.10.2016 um 11:57 schrieb H. Nikolaus Schaller <[email protected]>:

Hi,

Quoted 10 lines

Review

Quoted 7 lines

(quoted diff) +7 -0

+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

Quoted 2 lines

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

Sorry, but I think I do not completely understand what you mean here.

The problem is that we need to allocate some struct tsc2007 in both cases. But in one case managed directly by &client->dev and in the other managed indirectly. This is why I use the private area of struct iio_dev to store the full struct tsc2007 and not just a pointer.

Quoted 11 lines

Having doing that, you can have this CONFIG_IIO block as just doing the iio stuff with the input elements pulled back into the main probe function.

Then define something like

iio_configure (stubbed to nothing if no IIO) and
iio_unconfigure (also stubbed to nothing if no IIO).

Quoted 2 lines

A couple of additions in the header

Quoted 3 lines

to make it all work
(the struct tsc2007 and tsc2007_xfer() + a few of the register defines..

Hi Jonathan,

Quoted 19 lines

Am 23.10.2016 um 11:57 schrieb H. Nikolaus Schaller <[email protected]>:

Hi,

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev)

Quoted 7 lines

drivers/input/touchscreen/tsc2007.c +7 -0

+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

Quoted 2 lines

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

Sorry, but I think I do not completely understand what you mean here.

Quoted 4 lines

The problem is that we need to allocate some struct tsc2007 in both cases. But in one case managed directly by &client->dev and in the other managed indirectly. This is why I use the private area of struct iio_dev to store the full struct tsc2007 and not just a pointer.

Ok, I think I finally did understand how you mean this and have started to implement something.

The idea is to have one alloc function to return a struct tsc2007. This can be part of the probe function, like it is in the unpatched driver.

In case of iio this struct tsc2007 is also allocated explicitly so that a pointer can be stored in iio_priv.

This just means an additional iio_priv->ts = devm_kzalloc() in case of iio.

I have added that approach to my inlined patch and it seems to work (attached).

Sorry if I do not use the wording you would use and sometimes overlook something you have said. I feel here like moving on thin ice and doing guesswork about unspoken assumptions...

Quoted 7 lines

Quoted 7 lines

Having doing that, you can have this CONFIG_IIO block as just doing the iio stuff with the input elements pulled back into the main probe function.

Quoted 5 lines

Quoted 5 lines

Then define something like

iio_configure (stubbed to nothing if no IIO) and
iio_unconfigure (also stubbed to nothing if no IIO).

This seems to work (draft attached).

Quoted 2 lines

Quoted 2 lines

A couple of additions in the header

I think you mean tsc2007.h?

This currently contains only platform data and could IMHO be eliminated if everything becomes DT.

Quoted 3 lines

Quoted 3 lines

to make it all work
(the struct tsc2007 and tsc2007_xfer() + a few of the register defines..

Here it appears to me that I have to make a lot of so far private static and even static inline functions public so that I can make them stubs and call them from tsc2007_iio.c.

And for having proper parameter types I have to make most private structs also public.

I really like the idea to have the optional iio feature in a separate source file, but when really starting to write code, I get the impression that it introduces more problems than it solves.

And I wonder a little why it is not done for #ifdef CONFIG_OF in tsc2007.c as well. There are also two static function in some #ifdef #else # endif and not going through stubs.

So is this intended to give up some static definitions?

drivers/input/touchscreen/tsc2007.c +6 -0

        int                     (*get_pendown_state)(struct device *);
        void                    (*clear_penirq)(void);
+
+       struct mutex            mlock;
+       void                    *private;
 };
 
 static inline int tsc2007_xfer(struct tsc2007 *tsc, u8 cmd)

@@ -192,7 +196,10 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
        while (!ts->stopped && tsc2007_is_pen_down(ts)) {
 
                /* pen is down, continue with the measurement */
+
+               mutex_lock(&ts->mlock);
                tsc2007_read_values(ts, &tc);
+               mutex_unlock(&ts->mlock);
 
                rt = tsc2007_calculate_resistance(ts, &tc);
 

drivers/input/touchscreen/tsc2007.c +157 -1

+};
+
+#define TSC2007_CHAN_IIO(_chan, _name, _type, _chan_info) \
+{ \
+       .datasheet_name = _name, \
+       .type = _type, \
+       .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |  \
+                       BIT(_chan_info), \
+       .indexed = 1, \
+       .channel = _chan, \
+}
+
+static const struct iio_chan_spec tsc2007_iio_channel[] = {
+       TSC2007_CHAN_IIO(0, "x", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(1, "y", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(2, "z1", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(3, "z2", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(4, "adc", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(5, "rt", IIO_VOLTAGE, IIO_CHAN_INFO_RAW), /* Ohms? */
+       TSC2007_CHAN_IIO(6, "pen", IIO_PRESSURE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(7, "temp0", IIO_TEMP, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(8, "temp1", IIO_TEMP, IIO_CHAN_INFO_RAW),
+};
+
+static int tsc2007_read_raw(struct iio_dev *indio_dev,
+       struct iio_chan_spec const *chan, int *val, int *val2, long mask)
+{
+       struct tsc2007_iio *iio = iio_priv(indio_dev);
+       struct tsc2007 *tsc = iio->ts;
+       int adc_chan = chan->channel;
+       int ret = 0;
+
+       if (adc_chan >= ARRAY_SIZE(tsc2007_iio_channel))
+               return -EINVAL;
+
+       if (mask != IIO_CHAN_INFO_RAW)
+               return -EINVAL;
+
+       mutex_lock(&tsc->mlock);
+
+       switch (chan->channel) {
+       case 0:
+               *val = tsc2007_xfer(tsc, READ_X);
+               break;
+       case 1:
+               *val = tsc2007_xfer(tsc, READ_Y);
+               break;
+       case 2:
+               *val = tsc2007_xfer(tsc, READ_Z1);
+               break;
+       case 3:
+               *val = tsc2007_xfer(tsc, READ_Z2);
+               break;
+       case 4:
+               *val = tsc2007_xfer(tsc, (ADC_ON_12BIT | TSC2007_MEASURE_AUX));
+               break;
+       case 5: {
+               struct ts_event tc;
+
+               tc.x = tsc2007_xfer(tsc, READ_X);
+               tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+               tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+               *val = tsc2007_calculate_resistance(tsc, &tc);
+               break;
+       }
+       case 6:
+               *val = tsc2007_is_pen_down(tsc);
+               break;
+       case 7:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP0));
+               break;
+       case 8:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP1));
+               break;
+       }
+
+       /* Prepare for next touch reading - power down ADC, enable PENIRQ */
+       tsc2007_xfer(tsc, PWRDOWN);
+
+       mutex_unlock(&tsc->mlock);
+
+       ret = IIO_VAL_INT;
+
+       return ret;
+}
+
+static const struct iio_info tsc2007_iio_info = {
+       .read_raw = tsc2007_read_raw,
+       .driver_module = THIS_MODULE,
+};
+
+static inline int tsc2007_iio_configure(struct tsc2007 *ts)
+{
+       int err;
+       struct iio_dev *indio_dev;
+       struct tsc2007_iio *iio;
+
+       indio_dev = devm_iio_device_alloc(&ts->client->dev, sizeof(struct tsc2007_iio));
+       if (!indio_dev) {
+               dev_err(&ts->client->dev, "iio_device_alloc failed\n");
+               return -ENOMEM;
+       }
+
+       iio = iio_priv(indio_dev);
+       iio->ts = ts;
+       ts->private = (void *) indio_dev;
+
+       indio_dev->name = "tsc2007";
+       indio_dev->dev.parent = &ts->client->dev;
+       indio_dev->info = &tsc2007_iio_info;
+       indio_dev->modes = INDIO_DIRECT_MODE;
+       indio_dev->channels = tsc2007_iio_channel;
+       indio_dev->num_channels = ARRAY_SIZE(tsc2007_iio_channel);
+
+       err = iio_device_register(indio_dev);
+       if (err < 0) {
+               dev_err(&ts->client->dev, "iio_device_register() failed: %d\n",
+                       err);
+               return err;
+       }
+
+       return 0;
+}
+
+static inline void tsc2007_iio_unconfigure(struct tsc2007 *ts)
+{
+       struct iio_dev *indio_dev = ts->private;
+
+       iio_device_unregister(indio_dev);
+}
+
+#else /* CONFIG_IIO */
+
+static inline int tsc2007_iio_configure(struct tsc2007 *ts)
+{
+       /* not needed */
+}
+
+static inline void tsc2007_iio_unconfigure(struct tsc2007 *ts)
+{
+       /* not needed */
+}
+
+#endif /* CONFIG_IIO */
+
 #ifdef CONFIG_OF
 static int tsc2007_get_pendown_state_gpio(struct device *dev)
 {

@@ -472,7 +630,13 @@ static int tsc2007_probe(struct i2c_client *client,
        ts->client = client;
        ts->irq = client->irq;
        ts->input = input_dev;
+
+       err = tsc2007_iio_configure(ts);
+       if (err < 0)
+               return err;
+
        init_waitqueue_head(&ts->wait);
+       mutex_init(&ts->mlock);
 
        snprintf(ts->phys, sizeof(ts->phys),
                 "%s/input0", dev_name(&client->dev));

@@ -503,8 +667,10 @@ static int tsc2007_probe(struct i2c_client *client,
                                                          ts->fuzzz, 0);
        } else {
                err = tsc2007_probe_dt(client, ts);
-               if (err)
+               if (err) {
+                       tsc2007_iio_unconfigure(ts);
                        return err;
+               }
        }
 
        if (pdata) {

@@ -516,6 +682,7 @@ static int tsc2007_probe(struct i2c_client *client,
                                dev_err(&client->dev,
                                        "Failed to register exit_platform_hw action, %d\n",
                                        err);
+                               tsc2007_iio_unconfigure(ts);
                                return err;
                        }
                }

drivers/input/touchscreen/tsc2007.c +11 -0

                return err;
        }
 

@@ -543,6 +711,7 @@ static int tsc2007_probe(struct i2c_client *client,
        if (err < 0) {
                dev_err(&client->dev,
                        "Failed to setup chip: %d\n", err);
+               tsc2007_iio_unconfigure(ts);
                return err;     /* usually, chip does not respond */
        }
 

@@ -550,12 +719,21 @@ static int tsc2007_probe(struct i2c_client *client,
        if (err) {
                dev_err(&client->dev,
                        "Failed to register input device: %d\n", err);
+               tsc2007_iio_unconfigure(ts);
                return err;
        }
 
        return 0;
 }
 
+static int tsc2007_remove(struct i2c_client *client)
+{
+       struct tsc2007 *ts = i2c_get_clientdata(client);
+       tsc2007_iio_unconfigure(ts);
+       input_unregister_device(ts->input);
+       return 0;
+}
+
 static const struct i2c_device_id tsc2007_idtable[] = {
        { "tsc2007", 0 },
        { }

@@ -578,6 +756,7 @@ static struct i2c_driver tsc2007_driver = {
        },
        .id_table       = tsc2007_idtable,
        .probe          = tsc2007_probe,
+       .remove         = tsc2007_remove,
 };
 
 module_i2c_driver(tsc2007_driver);--

To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Am 23.10.2016 um 21:00 schrieb Jonathan Cameron <[email protected]>:

Quoted 21 lines · H. Nikolaus Schaller

Hi Jonathan,

Quoted 19 lines

Am 23.10.2016 um 11:57 schrieb H. Nikolaus Schaller <[email protected]>:

Hi,

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev)

Quoted 7 lines

drivers/input/touchscreen/tsc2007.c +7 -0

+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

Quoted 2 lines

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

Sorry, but I think I do not completely understand what you mean here.

Quoted 8 lines

Quoted 4 lines

The problem is that we need to allocate some struct tsc2007 in both cases. But in one case managed directly by &client->dev and in the other managed indirectly. This is why I use the private area of struct iio_dev to store the full struct tsc2007 and not just a pointer.

Ok, I think I finally did understand how you mean this and have started to implement something.

oops. Didn't look on in my emails to get to this one!

Quoted 6 lines

The idea is to have one alloc function to return a struct tsc2007. This can be part of the probe function, like it is in the unpatched driver.

In case of iio this struct tsc2007 is also allocated explicitly so that a pointer can be stored in iio_priv.

Quoted 7 lines

This just means an additional iio_priv->ts = devm_kzalloc() in case of iio.

I have added that approach to my inlined patch and it seems to work (attached).

Sorry if I do not use the wording you would use and sometimes overlook something you have said. I feel here like moving on thin ice and doing guesswork about unspoken assumptions...

That's fine. Stuff that can appear obvious to one person is not necessarily obvious to another!

Quoted 13 lines

Quoted 12 lines

Quoted 12 lines

Having doing that, you can have this CONFIG_IIO block as just doing the iio stuff with the input elements pulled back into the main probe function.

Then define something like

iio_configure (stubbed to nothing if no IIO) and
iio_unconfigure (also stubbed to nothing if no IIO).

Quoted 6 lines

This seems to work (draft attached).

Quoted 2 lines

Quoted 2 lines

A couple of additions in the header

I think you mean tsc2007.h?

Nope. A local header alongside the driver is what you want for this stuff. driver/input/tsc2007.h

Quoted 11 lines

This currently contains only platform data and could IMHO be eliminated if everything becomes DT.

Quoted 3 lines

Quoted 3 lines

to make it all work
(the struct tsc2007 and tsc2007_xfer() + a few of the register defines..

Here it appears to me that I have to make a lot of so far private static and even static inline functions public so that I can make them stubs and call them from tsc2007_iio.c.

There will be a few.

Quoted 3 lines

And for having proper parameter types I have to make most private structs also public.

Yes a few of those as well.

Quoted 8 lines

I really like the idea to have the optional iio feature in a separate source file, but when really starting to write code, I get the impression that it introduces more problems than it solves.

And I wonder a little why it is not done for #ifdef CONFIG_OF in tsc2007.c as well. There are also two static function in some #ifdef #else # endif and not going through stubs.

Usually it is only done once a certain volume of code exists.

Quoted 2 lines

So is this intended to give up some static definitions?

Yes, that happens the moment you have multiple source files.

Some losses but generally end up with clean code separation. Always a trade off unfortunately. Pity we can't just insist IIO is available! Rather large to pull in for what is probable a niche use case.

Below is definitely heading in the right direction. I remember vaguely being convinced of the worth of doing this when optional code is involved! (was a good while ago now)

Jonathan

drivers/input/touchscreen/tsc2007.c line 33

Quoted 13 lines

BR and thanks,
Nikolaus

drivers/input/touchscreen/tsc2007.c +1 -0

index 5e3c4bf..92da8f6 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>

Should not need this after introducing the new file. Will only be needed in the iio specific .c file.

drivers/input/touchscreen/tsc2007.c struct tsc2007 {

Quoted 4 lines

drivers/input/touchscreen/tsc2007.c +0 -0

 #define TSC2007_MEASURE_TEMP0          (0x0 << 4)
 #define TSC2007_MEASURE_AUX            (0x2 << 4)

@@ -98,6 +99,9 @@ struct tsc2007 {

This will definitely need to go in the header though.

Quoted 21 lines

drivers/input/touchscreen/tsc2007.c +6 -0

       int                     (*get_pendown_state)(struct device *);
       void                    (*clear_penirq)(void);
+
+       struct mutex            mlock;
+       void                    *private;
 };
 
 static inline int tsc2007_xfer(struct tsc2007 *tsc, u8 cmd)

@@ -192,7 +196,10 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
       while (!ts->stopped && tsc2007_is_pen_down(ts)) {
 
               /* pen is down, continue with the measurement */
+
+               mutex_lock(&ts->mlock);
               tsc2007_read_values(ts, &tc);
+               mutex_unlock(&ts->mlock);
 
               rt = tsc2007_calculate_resistance(ts, &tc);
 

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev) · line 332

Quoted 7 lines

drivers/input/touchscreen/tsc2007.c +4 -0

@@ -319,6 +326,157 @@ static void tsc2007_close(struct input_dev *input_dev)
       tsc2007_stop(ts);
 }
 
+#ifdef CONFIG_IIO
+
+struct tsc2007_iio {
+       struct tsc2007 *ts;

Spot on.

Quoted 175 lines

drivers/input/touchscreen/tsc2007.c +156 -1

+};
+
+#define TSC2007_CHAN_IIO(_chan, _name, _type, _chan_info) \
+{ \
+       .datasheet_name = _name, \
+       .type = _type, \
+       .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |  \
+                       BIT(_chan_info), \
+       .indexed = 1, \
+       .channel = _chan, \
+}
+
+static const struct iio_chan_spec tsc2007_iio_channel[] = {
+       TSC2007_CHAN_IIO(0, "x", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(1, "y", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(2, "z1", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(3, "z2", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(4, "adc", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(5, "rt", IIO_VOLTAGE, IIO_CHAN_INFO_RAW), /* Ohms? */
+       TSC2007_CHAN_IIO(6, "pen", IIO_PRESSURE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(7, "temp0", IIO_TEMP, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(8, "temp1", IIO_TEMP, IIO_CHAN_INFO_RAW),
+};
+
+static int tsc2007_read_raw(struct iio_dev *indio_dev,
+       struct iio_chan_spec const *chan, int *val, int *val2, long mask)
+{
+       struct tsc2007_iio *iio = iio_priv(indio_dev);
+       struct tsc2007 *tsc = iio->ts;
+       int adc_chan = chan->channel;
+       int ret = 0;
+
+       if (adc_chan >= ARRAY_SIZE(tsc2007_iio_channel))
+               return -EINVAL;
+
+       if (mask != IIO_CHAN_INFO_RAW)
+               return -EINVAL;
+
+       mutex_lock(&tsc->mlock);
+
+       switch (chan->channel) {
+       case 0:
+               *val = tsc2007_xfer(tsc, READ_X);
+               break;
+       case 1:
+               *val = tsc2007_xfer(tsc, READ_Y);
+               break;
+       case 2:
+               *val = tsc2007_xfer(tsc, READ_Z1);
+               break;
+       case 3:
+               *val = tsc2007_xfer(tsc, READ_Z2);
+               break;
+       case 4:
+               *val = tsc2007_xfer(tsc, (ADC_ON_12BIT | TSC2007_MEASURE_AUX));
+               break;
+       case 5: {
+               struct ts_event tc;
+
+               tc.x = tsc2007_xfer(tsc, READ_X);
+               tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+               tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+               *val = tsc2007_calculate_resistance(tsc, &tc);
+               break;
+       }
+       case 6:
+               *val = tsc2007_is_pen_down(tsc);
+               break;
+       case 7:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP0));
+               break;
+       case 8:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP1));
+               break;
+       }
+
+       /* Prepare for next touch reading - power down ADC, enable PENIRQ */
+       tsc2007_xfer(tsc, PWRDOWN);
+
+       mutex_unlock(&tsc->mlock);
+
+       ret = IIO_VAL_INT;
+
+       return ret;
+}
+
+static const struct iio_info tsc2007_iio_info = {
+       .read_raw = tsc2007_read_raw,
+       .driver_module = THIS_MODULE,
+};
+
+static inline int tsc2007_iio_configure(struct tsc2007 *ts)
+{
+       int err;
+       struct iio_dev *indio_dev;
+       struct tsc2007_iio *iio;
+
+       indio_dev = devm_iio_device_alloc(&ts->client->dev, sizeof(struct tsc2007_iio));
+       if (!indio_dev) {
+               dev_err(&ts->client->dev, "iio_device_alloc failed\n");
+               return -ENOMEM;
+       }
+
+       iio = iio_priv(indio_dev);
+       iio->ts = ts;
+       ts->private = (void *) indio_dev;
+
+       indio_dev->name = "tsc2007";
+       indio_dev->dev.parent = &ts->client->dev;
+       indio_dev->info = &tsc2007_iio_info;
+       indio_dev->modes = INDIO_DIRECT_MODE;
+       indio_dev->channels = tsc2007_iio_channel;
+       indio_dev->num_channels = ARRAY_SIZE(tsc2007_iio_channel);
+
+       err = iio_device_register(indio_dev);
+       if (err < 0) {
+               dev_err(&ts->client->dev, "iio_device_register() failed: %d\n",
+                       err);
+               return err;
+       }
+
+       return 0;
+}
+
+static inline void tsc2007_iio_unconfigure(struct tsc2007 *ts)
+{
+       struct iio_dev *indio_dev = ts->private;
+
+       iio_device_unregister(indio_dev);
+}
+
+#else /* CONFIG_IIO */
+
+static inline int tsc2007_iio_configure(struct tsc2007 *ts)
+{
+       /* not needed */
+}
+
+static inline void tsc2007_iio_unconfigure(struct tsc2007 *ts)
+{
+       /* not needed */
+}
+
+#endif /* CONFIG_IIO */
+
 #ifdef CONFIG_OF
 static int tsc2007_get_pendown_state_gpio(struct device *dev)
 {

@@ -472,7 +630,13 @@ static int tsc2007_probe(struct i2c_client *client,
       ts->client = client;
       ts->irq = client->irq;
       ts->input = input_dev;
+
+       err = tsc2007_iio_configure(ts);
+       if (err < 0)
+               return err;
+
       init_waitqueue_head(&ts->wait);
+       mutex_init(&ts->mlock);
 
       snprintf(ts->phys, sizeof(ts->phys),
                "%s/input0", dev_name(&client->dev));

@@ -503,8 +667,10 @@ static int tsc2007_probe(struct i2c_client *client,
                                                         ts->fuzzz, 0);
       } else {
               err = tsc2007_probe_dt(client, ts);
-               if (err)
+               if (err) {
+                       tsc2007_iio_unconfigure(ts);
                       return err;
+               }
       }
 

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c +1 -0

       if (pdata) {

@@ -516,6 +682,7 @@ static int tsc2007_probe(struct i2c_client *client,
                               dev_err(&client->dev,
                                       "Failed to register exit_platform_hw action, %d\n",
                                       err);
+                               tsc2007_iio_unconfigure(ts);
                               return err;
                       }
               }

drivers/input/touchscreen/tsc2007.c static int tsc2007_probe(struct i2c_client *client, · line 703

Quoted 4 lines

drivers/input/touchscreen/tsc2007.c +1 -0

@@ -533,6 +700,7 @@ static int tsc2007_probe(struct i2c_client *client,
       if (err) {
               dev_err(&client->dev, "Failed to request irq %d: %d\n",
                       ts->irq, err);
+               tsc2007_iio_unconfigure(ts);

Maybe need a common unwind and use a goto to get to it.

Quoted 40 lines

drivers/input/touchscreen/tsc2007.c +11 -0

               return err;
       }
 

@@ -543,6 +711,7 @@ static int tsc2007_probe(struct i2c_client *client,
       if (err < 0) {
               dev_err(&client->dev,
                       "Failed to setup chip: %d\n", err);
+               tsc2007_iio_unconfigure(ts);
               return err;     /* usually, chip does not respond */
       }
 

@@ -550,12 +719,21 @@ static int tsc2007_probe(struct i2c_client *client,
       if (err) {
               dev_err(&client->dev,
                       "Failed to register input device: %d\n", err);
+               tsc2007_iio_unconfigure(ts);
               return err;
       }
 
       return 0;
 }
 
+static int tsc2007_remove(struct i2c_client *client)
+{
+       struct tsc2007 *ts = i2c_get_clientdata(client);
+       tsc2007_iio_unconfigure(ts);
+       input_unregister_device(ts->input);
+       return 0;
+}
+
 static const struct i2c_device_id tsc2007_idtable[] = {
       { "tsc2007", 0 },
       { }

@@ -578,6 +756,7 @@ static struct i2c_driver tsc2007_driver = {
       },
       .id_table       = tsc2007_idtable,
       .probe          = tsc2007_probe,
+       .remove         = tsc2007_remove,
 };
 

Quoted 4 lines

module_i2c_driver(tsc2007_driver);--
To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Am 23.10.2016 um 21:00 schrieb Jonathan Cameron <[email protected]>:

Quoted 21 lines · H. Nikolaus Schaller

Hi Jonathan,

Quoted 19 lines

Am 23.10.2016 um 11:57 schrieb H. Nikolaus Schaller <[email protected]>:

Hi,

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev)

Quoted 7 lines

drivers/input/touchscreen/tsc2007.c +7 -0

+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

Quoted 2 lines

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

Sorry, but I think I do not completely understand what you mean here.

Quoted 8 lines

Quoted 4 lines

The problem is that we need to allocate some struct tsc2007 in both cases. But in one case managed directly by &client->dev and in the other managed indirectly. This is why I use the private area of struct iio_dev to store the full struct tsc2007 and not just a pointer.

Ok, I think I finally did understand how you mean this and have started to implement something.

oops. Didn't look on in my emails to get to this one!

Quoted 6 lines

The idea is to have one alloc function to return a struct tsc2007. This can be part of the probe function, like it is in the unpatched driver.

In case of iio this struct tsc2007 is also allocated explicitly so that a pointer can be stored in iio_priv.

Quoted 7 lines

This just means an additional iio_priv->ts = devm_kzalloc() in case of iio.

I have added that approach to my inlined patch and it seems to work (attached).

Sorry if I do not use the wording you would use and sometimes overlook something you have said. I feel here like moving on thin ice and doing guesswork about unspoken assumptions...

That's fine. Stuff that can appear obvious to one person is not necessarily obvious to another!

Quoted 13 lines

Quoted 7 lines

Quoted 7 lines

Having doing that, you can have this CONFIG_IIO block as just doing the iio stuff with the input elements pulled back into the main probe function.

Quoted 5 lines

Quoted 5 lines

Then define something like

iio_configure (stubbed to nothing if no IIO) and
iio_unconfigure (also stubbed to nothing if no IIO).

Quoted 6 lines

This seems to work (draft attached).

Quoted 2 lines

Quoted 2 lines

A couple of additions in the header

I think you mean tsc2007.h?

Nope. A local header alongside the driver is what you want for this stuff. driver/input/tsc2007.h

Quoted 11 lines

This currently contains only platform data and could IMHO be eliminated if everything becomes DT.

Quoted 3 lines

Quoted 3 lines

to make it all work
(the struct tsc2007 and tsc2007_xfer() + a few of the register defines..

Here it appears to me that I have to make a lot of so far private static and even static inline functions public so that I can make them stubs and call them from tsc2007_iio.c.

There will be a few.

Quoted 3 lines

And for having proper parameter types I have to make most private structs also public.

Yes a few of those as well.

Quoted 8 lines

I really like the idea to have the optional iio feature in a separate source file, but when really starting to write code, I get the impression that it introduces more problems than it solves.

And I wonder a little why it is not done for #ifdef CONFIG_OF in tsc2007.c as well. There are also two static function in some #ifdef #else # endif and not going through stubs.

Usually it is only done once a certain volume of code exists.

Quoted 2 lines

So is this intended to give up some static definitions?

Yes, that happens the moment you have multiple source files.

Some losses but generally end up with clean code separation. Always a trade off unfortunately. Pity we can't just insist IIO is available! Rather large to pull in for what is probable a niche use case.

Below is definitely heading in the right direction. I remember vaguely being convinced of the worth of doing this when optional code is involved! (was a good while ago now)

Jonathan

drivers/input/touchscreen/tsc2007.c line 33

Quoted 13 lines

BR and thanks,
Nikolaus

drivers/input/touchscreen/tsc2007.c +1 -0

index 5e3c4bf..92da8f6 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>

Should not need this after introducing the new file. Will only be needed in the iio specific .c file.

Quoted 1 line

drivers/input/touchscreen/tsc2007.c struct tsc2007 {

Quoted 3 lines

#define TSC2007_MEASURE_TEMP0 (0x0 << 4)
#define TSC2007_MEASURE_AUX (0x2 << 4)

drivers/input/touchscreen/tsc2007.c +0 -0

@@ -98,6 +99,9 @@ struct tsc2007 {

This will definitely need to go in the header though.

Hi Jonathan,

Quoted 123 lines

Am 23.10.2016 um 21:00 schrieb Jonathan Cameron <[email protected]>:

Quoted 21 lines · H. Nikolaus Schaller

Hi Jonathan,

Quoted 19 lines

Am 23.10.2016 um 11:57 schrieb H. Nikolaus Schaller <[email protected]>:

Hi,

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev)

Quoted 7 lines

drivers/input/touchscreen/tsc2007.c +7 -0

+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

Quoted 2 lines

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

Sorry, but I think I do not completely understand what you mean here.

Quoted 8 lines

Quoted 4 lines

The problem is that we need to allocate some struct tsc2007 in both cases. But in one case managed directly by &client->dev and in the other managed indirectly. This is why I use the private area of struct iio_dev to store the full struct tsc2007 and not just a pointer.

Ok, I think I finally did understand how you mean this and have started to implement something.

oops. Didn't look on in my emails to get to this one!

Quoted 6 lines

The idea is to have one alloc function to return a struct tsc2007. This can be part of the probe function, like it is in the unpatched driver.

In case of iio this struct tsc2007 is also allocated explicitly so that a pointer can be stored in iio_priv.

Quoted 7 lines

This just means an additional iio_priv->ts = devm_kzalloc() in case of iio.

I have added that approach to my inlined patch and it seems to work (attached).

Sorry if I do not use the wording you would use and sometimes overlook something you have said. I feel here like moving on thin ice and doing guesswork about unspoken assumptions...

That's fine. Stuff that can appear obvious to one person is not necessarily obvious to another!

Quoted 13 lines

Quoted 7 lines

Quoted 7 lines

Having doing that, you can have this CONFIG_IIO block as just doing the iio stuff with the input elements pulled back into the main probe function.

Quoted 5 lines

Quoted 5 lines

Then define something like

iio_configure (stubbed to nothing if no IIO) and
iio_unconfigure (also stubbed to nothing if no IIO).

Quoted 6 lines

This seems to work (draft attached).

Quoted 2 lines

Quoted 2 lines

A couple of additions in the header

I think you mean tsc2007.h?

Nope. A local header alongside the driver is what you want for this stuff. driver/input/tsc2007.h

Quoted 11 lines

This currently contains only platform data and could IMHO be eliminated if everything becomes DT.

Quoted 3 lines

Quoted 3 lines

to make it all work
(the struct tsc2007 and tsc2007_xfer() + a few of the register defines..

Here it appears to me that I have to make a lot of so far private static and even static inline functions public so that I can make them stubs and call them from tsc2007_iio.c.

There will be a few.

Quoted 3 lines

And for having proper parameter types I have to make most private structs also public.

Yes a few of those as well.

Quoted 8 lines

I really like the idea to have the optional iio feature in a separate source file, but when really starting to write code, I get the impression that it introduces more problems than it solves.

And I wonder a little why it is not done for #ifdef CONFIG_OF in tsc2007.c as well. There are also two static function in some #ifdef #else # endif and not going through stubs.

Usually it is only done once a certain volume of code exists.

Quoted 2 lines

So is this intended to give up some static definitions?

Yes, that happens the moment you have multiple source files.

Some losses but generally end up with clean code separation. Always a trade off unfortunately. Pity we can't just insist IIO is available! Rather large to pull in for what is probable a niche use case.

Below is definitely heading in the right direction. I remember vaguely being convinced of the worth of doing this when optional code is involved! (was a good while ago now)

Jonathan

drivers/input/touchscreen/tsc2007.c line 33

Quoted 13 lines

BR and thanks,
Nikolaus

drivers/input/touchscreen/tsc2007.c +1 -0

index 5e3c4bf..92da8f6 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>

Should not need this after introducing the new file. Will only be needed in the iio specific .c file.

Quoted 1 line

Quoted 4 lines

drivers/input/touchscreen/tsc2007.c struct tsc2007 {

Quoted 3 lines

#define TSC2007_MEASURE_TEMP0 (0x0 << 4)
#define TSC2007_MEASURE_AUX (0x2 << 4)

drivers/input/touchscreen/tsc2007.c +0 -0

@@ -98,6 +99,9 @@ struct tsc2007 {

This will definitely need to go in the header though.

Now I have split the code into:

tsc2007.h (constants, structs and stubs)
tsc2007_iio.c (the iio stuff)
tsc2007.c (most parts of the original driver)

but I have a problem of correctly modifying the Makefile.

It currently looks like:

obj-$(CONFIG_TOUCHSCREEN_TSC2007) += tsc2007.o
obj-$(CONFIG_IIO) += tsc2007_iio.o

We have configured CONFIG_TOUCHSCREEN_TSC2007=m and CONFIG_IIO=y.

This obviously compiles tsc2007_iio.o into the kernel.

This means that tsc2007_iio.o references tsc2007_xfer which is part of the module.

I would like to get both linked into the module, but the iio part obviously only if CONFIG_IIO is defined (either -y or -m).

How can I define this?

Or can I define

obj-$(CONFIG_TOUCHSCREEN_TSC2007) += tsc2007.o tsc2007_iio.o

and embrace all code in tsc2007_iio with a big #ifdef CONFIG_IIO so that it is compiled into an empty object file in the non-iio case?

BR and thanks,
Nikolaus

--
To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Hi,

Quoted 53 lines

Am 23.10.2016 um 11:24 schrieb Jonathan Cameron <[email protected]>:

Quoted 50 lines · H. Nikolaus Schaller

Hi Jonathan,

Quoted 17 lines

Am 22.10.2016 um 20:33 schrieb Jonathan Cameron <[email protected]>:

Quoted 11 lines · H. Nikolaus Schaller

The tsc2007 chip not only has a resistive touch screen controller but also an external AUX adc imput which can be used for an ambient light sensor, battery voltage monitoring or any general purpose.

Additionally it can measure the chip temperature.

This extension provides an iio interface for these adc channels.

Since it is not wasting much resources and is very straightforward, we simply provide all other adc channels as optional iio interfaces as weel. This can be used for debugging or special applications.

well

Quoted 2 lines

Signed-off-by H. Nikolaus Schaller <[email protected]>

Quoted 8 lines

This could be cleaner done perhaps by factoring out the IIO stuff into a separate file and using a header with stubs to deal with the no available case.

There will only be a handful of stubs and it'll give you a lot cleaner code in here.

If def fun in .c files is always harder to deal with than in a header where stubs are really obvious.

Yes, it became a lot of #ifdefs spread over the source file.

The easiest thing would be to require IIO to be enabled :)

With your proposal to consider refactoring, I think the crucial part is the conditional allocation either through devm_iio_device_alloc() or devm_kzalloc(). This can be refactored into some conditional tsc2007_alloc().

I have tried some draft (not tested and not tidied up) to check if the direction is good.

This reduces the number of #ifdef CONFIG_IIO from 7 to 2 without introducing new files or includes. There are also 2 other #ifdef CONFIG_OF so it doesn't seem to be very complex now in comparison. And the patch itself has only a handful of hunks (8).

Moving tsc2007_alloc into a separate file tsc2007_iio.c would only move around one #ifdef CONFIG_OF from tsc2007.c but IMHO makes it more difficult to understand because it is not really iio specific and one has to switch between two source files. And I would have to touch the Makefile as well.

Quoted 6 lines

Quoted 1 line

What do you think?

I'd still split it. The only bit of the IIO block that isn't specific is a tiny chunk of the allocation code (as you've highlighted).

Even that can be avoided by adding a tiny bit more indirection than would otherwise be needed (it's not pretty but it would give a clean separation).

I hope I understand what you mean (which is an indication that the result may be much easier to read for you but not me...).

Quoted 27 lines

It's pretty much the way this sort of optional functionality should always be done - means that if you don't care (i.e. it's not enabled) you don't even have to see the code.

Jonathan

Quoted 21 lines

If generally ok, I can include that in [PATCH v5].

BR and thanks,
Nikolaus

drivers/input/touchscreen/tsc2007.c +1 -0

index 5e3c4bf..691e79f 100644
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/input/touchscreen.h>
+#include <linux/iio/iio.h>
 
 #define TSC2007_MEASURE_TEMP0          (0x0 << 4)
 #define TSC2007_MEASURE_AUX            (0x2 << 4)

@@ -69,9 +70,13 @@ struct ts_event {
 

Quoted 9 lines

drivers/input/touchscreen/tsc2007.c struct ts_event { · line 75

Quoted 5 lines

drivers/input/touchscreen/tsc2007.c +3 -0

 struct tsc2007 {
       struct input_dev        *input;
+#ifdef CONFIG_IIO
+       struct iio_dev          *indio;
+#endif

I wouldn't bother with this one. Just have struct iio_dev; before this and it'll waste a whole one pointer (+ you shouldn't need to have iio.h included in here once you have spit the files).

Looks as if I have to make a knot in my brain before I start to understand...

How can I use struct iio_dev here w/o including iio.h?

Quoted 125 lines

Quoted 112 lines

drivers/input/touchscreen/tsc2007.c +93 -0

       char                    phys[32];
 
       struct i2c_client       *client;
+       struct mutex            mlock;
 
       u16                     model;
       u16                     x_plate_ohms;

@@ -192,7 +197,10 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
       while (!ts->stopped && tsc2007_is_pen_down(ts)) {
 
               /* pen is down, continue with the measurement */
+
+               mutex_lock(&ts->mlock);
               tsc2007_read_values(ts, &tc);
+               mutex_unlock(&ts->mlock);
 
               rt = tsc2007_calculate_resistance(ts, &tc);
 

@@ -319,6 +327,162 @@ static void tsc2007_close(struct input_dev *input_dev)
       tsc2007_stop(ts);
 }
 
+#ifdef CONFIG_IIO
+
+#define TSC2007_CHAN_IIO(_chan, _name, _type, _chan_info) \
+{ \
+       .datasheet_name = _name, \
+       .type = _type, \
+       .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |  \
+                       BIT(_chan_info), \
+       .indexed = 1, \
+       .channel = _chan, \
+}
+
+static const struct iio_chan_spec tsc2007_iio_channel[] = {
+       TSC2007_CHAN_IIO(0, "x", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(1, "y", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(2, "z1", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(3, "z2", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(4, "adc", IIO_VOLTAGE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(5, "rt", IIO_VOLTAGE, IIO_CHAN_INFO_RAW), /* Ohms? */
+       TSC2007_CHAN_IIO(6, "pen", IIO_PRESSURE, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(7, "temp0", IIO_TEMP, IIO_CHAN_INFO_RAW),
+       TSC2007_CHAN_IIO(8, "temp1", IIO_TEMP, IIO_CHAN_INFO_RAW),
+};
+
+static int tsc2007_read_raw(struct iio_dev *indio_dev,
+       struct iio_chan_spec const *chan, int *val, int *val2, long mask)
+{
+       struct  tsc2007 *tsc = iio_priv(indio_dev);
+       int adc_chan = chan->channel;
+       int ret = 0;
+
+       if (adc_chan >= ARRAY_SIZE(tsc2007_iio_channel))
+               return -EINVAL;
+
+       if (mask != IIO_CHAN_INFO_RAW)
+               return -EINVAL;
+
+       mutex_lock(&tsc->mlock);
+
+       switch (chan->channel) {
+       case 0:
+               *val = tsc2007_xfer(tsc, READ_X);
+               break;
+       case 1:
+               *val = tsc2007_xfer(tsc, READ_Y);
+               break;
+       case 2:
+               *val = tsc2007_xfer(tsc, READ_Z1);
+               break;
+       case 3:
+               *val = tsc2007_xfer(tsc, READ_Z2);
+               break;
+       case 4:
+               *val = tsc2007_xfer(tsc, (ADC_ON_12BIT | TSC2007_MEASURE_AUX));
+               break;
+       case 5: {
+               struct ts_event tc;
+
+               tc.x = tsc2007_xfer(tsc, READ_X);
+               tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+               tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+               *val = tsc2007_calculate_resistance(tsc, &tc);
+               break;
+       }
+       case 6:
+               *val = tsc2007_is_pen_down(tsc);
+               break;
+       case 7:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP0));
+               break;
+       case 8:
+               *val = tsc2007_xfer(tsc,
+                                   (ADC_ON_12BIT | TSC2007_MEASURE_TEMP1));
+               break;
+       }
+
+       /* Prepare for next touch reading - power down ADC, enable PENIRQ */
+       tsc2007_xfer(tsc, PWRDOWN);
+
+       mutex_unlock(&tsc->mlock);
+
+       ret = IIO_VAL_INT;
+
+       return ret;
+}
+
+static const struct iio_info tsc2007_iio_info = {
+       .read_raw = tsc2007_read_raw,

drivers/input/touchscreen/tsc2007.c static void tsc2007_close(struct input_dev *input_dev) · line 428

Quoted 10 lines

drivers/input/touchscreen/tsc2007.c +10 -0

+       .driver_module = THIS_MODULE,
+};
+
+static int tsc2007_alloc(struct i2c_client *client, struct tsc2007 **ts,
+                          struct input_dev **input_dev)
+{
+       int err;
+       struct iio_dev *indio_dev;
+
+       indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*ts));

Instead of doing this to reduce the delta between versions make iio_priv a struct tsc2007 **

Quoted 2 lines

That is have a single pointer in there and do your allocation of struct tsc2007 separately.

Sorry, but I think I do not completely understand what you mean here.

The problem is that we need to allocate some struct tsc2007 in both cases. But in one case managed directly by &client->dev and in the other managed indirectly. This is why I use the private area of struct iio_dev to store the full struct tsc2007 and not just a pointer.