input, touch, atmel_mxt_ts: use atmel driver

try to bring the driver from
https://github.com/ndyer/linux for-next

to mainline. Post this as a RFC, to see if this is possible...

based on:
commit dda0a5570574cfe467c1c794cf7a17e1d742ec02 ("Input:
atmel_mxt_ts - implement improved debug message interface")

from Nick Dyer <[email protected]>

added:
- remove some (not all) checkpatch warnings
- Add support for voltage regulator
- Add support for Device Tree support

---
 .../bindings/input/touchscreen/atmel_mxt_ts.txt    |   27 +
 arch/arm/mach-s5pv210/mach-goni.c                  |   27 +
 drivers/input/touchscreen/atmel_mxt_ts.c           | 2666 ++++++++++++++++----
 include/linux/i2c/atmel_mxt_ts.h                   |    8 +-
 4 files changed, 2245 insertions(+), 483 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/input/touchscreen/atmel_mxt_ts.txt

new file mode 100644
index 0000000..e025169
@@ -0,0 +1,27 @@
+* Atmel maXtouch touchscreen controller
+
+Required properties:
+- compatible: must be "atmel,maxtouch"
+- reg: I2C address of the chip
+- interrupt-parent: interrupt controller which provides the interrupt
+- interrupts: interrupt signal to which the chip is connected
+
+Optional properties:
+- vdd-supply: voltage regulator used for power control and reading
+  operating voltage
+
+Example:
+
+	i2c at 00000000 {
+		/* ... */
+
+		tsp at 4a {
+			compatible = "atmel,maxtouch";
+			reg = <0x4a>;
+			interrupt-parent = <&gpe1>;
+			interrupts = <7 2>;
+			vdd-supply = <&tsp_reg>;
+		};
+
+		/* ... */
+	};

index c1ce921..07d5006 100644
@@ -233,6 +233,32 @@ static void __init goni_radio_init(void)
 }
 
 /* TSP */
+static struct regulator_consumer_supply tsp_fixed_consumer =
+	REGULATOR_SUPPLY("vdd", "2-004a");
+
+static struct regulator_init_data tsp_fixed_voltage_init_data = {
+	.constraints		= {
+		.name		= "TSP_2.8V",
+	},
+	.num_consumer_supplies	= 1,
+	.consumer_supplies	= &tsp_fixed_consumer,
+};
+
+static struct fixed_voltage_config tsp_fixed_voltage_config = {
+	.supply_name		= "TSP_VDD",
+	.microvolts		= 2800000,
+	.gpio			= -EINVAL,
+	.init_data		= &tsp_fixed_voltage_init_data,
+};
+
+static struct platform_device tsp_fixed_voltage = {
+	.name			= "reg-fixed-voltage",
+	.id			= 3,
+	.dev			= {
+		.platform_data	= &tsp_fixed_voltage_config,
+	},
+};
+
 static struct mxt_platform_data qt602240_platform_data = {
 	.irqflags	= IRQF_TRIGGER_FALLING,
 };

@@ -832,6 +858,7 @@ static struct platform_device *goni_devices[] __initdata = {
 	&s3c_device_usb_hsotg,
 	&samsung_device_keypad,
 	&s3c_device_i2c1,
+	&tsp_fixed_voltage,
 	&s3c_device_i2c2,
 	&wm8994_fixed_voltage0,
 	&wm8994_fixed_voltage1,

index 3891dc1..bc74c3f 100644

1.8.3.1

Part of the diff is not shown here. View the raw message.

My thinking has been that the maintainers will more likely accept a series of small patches than to make such a huge change to the entire driver, my for-next tree is where I'm working on this, and I plan to post another set of 15 patches soon, when I've fixed up some issues to do with probe/remove cycles.

You should note that my for-next branch has moved on slightly since you have taken this, because I have implemented device tree support and changed the config download to use the request_firmware_async() function.

Your device tree stuff for mach-goni.c looks really useful, however - if you can send me a patch for that against my current for-next branch I can include it with my version of the device tree support.

all the best

Nick

Hello Nick,

Am 11.06.2014 14:06, schrieb Nick Dyer:

Sound good!

Oh, I take a look at this, thanks!

Ok, I send you this ASAP.

Thanks for this info.

bye,
Heiko

DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany