Skip to contentSkip to conversation

input: Added TSC2003

PATCH RESEND In review Richard Röjfors 4 participants 7 messages Jun 15, 2009 – Jun 19, 2009

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH RESEND 1/1 input: Added TSC2003

Supplied is a driver for the TSC2003. There is actually a driver for TSC2007 which can be used in some cases. The platform struct is reused from the TSC2007.
There is a big difference in the implementation between the drivers, this one does not use HR timers. The TSC2007 driver performs synchronous I2C in the timer callback (SW IRQ context) which is bad when the I2C driver is interrupt driven.

---

drivers/input/touchscreen/tsc2003.c +387 -0
Index: linux-2.6.30-rc7/drivers/input/touchscreen/tsc2003.c
===================================================================
@@ -0,0 +1,387 @@
+/*
+ * tsc2003.c Driver for TI TSC2003 touch screen controller
+ * Copyright (c) 2009 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/* Supports:
+ * TI TSC2003
+ *
+ * Inspired by tsc2007, Copyright (c) 2008 MtekVision Co., Ltd.
+ */
+#include <linux/module.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/i2c.h>
+#include <linux/i2c/tsc2007.h>
+#include <linux/kthread.h>
+#include <linux/semaphore.h>
+
+#define TSC2003_DRIVER_NAME		"tsc2003"
+
+#define TS_POLL_PERIOD	20		/* ms delay between samples */
+
+#define TSC2003_MEASURE_TEMP0		(0x0 << 4)
+#define TSC2003_MEASURE_AUX		(0x2 << 4)
+#define TSC2003_MEASURE_TEMP1		(0x4 << 4)
+#define TSC2003_ACTIVATE_XN		(0x8 << 4)
+#define TSC2003_ACTIVATE_YN		(0x9 << 4)
+#define TSC2003_ACTIVATE_YP_XN		(0xa << 4)
+#define TSC2003_SETUP			(0xb << 4)
+#define TSC2003_MEASURE_X		(0xc << 4)
+#define TSC2003_MEASURE_Y		(0xd << 4)
+#define TSC2003_MEASURE_Z1		(0xe << 4)
+#define TSC2003_MEASURE_Z2		(0xf << 4)
+
+#define TSC2003_POWER_OFF_IRQ_EN	(0x0 << 2)
+#define TSC2003_ADC_ON_IRQ_DIS0		(0x1 << 2)
+#define TSC2003_ADC_OFF_IRQ_EN		(0x2 << 2)
+#define TSC2003_ADC_ON_IRQ_DIS1		(0x3 << 2)
+
+#define TSC2003_12BIT			(0x0 << 1)
+#define TSC2003_8BIT			(0x1 << 1)
+
+#define	MAX_12BIT			((1 << 12) - 1)
+
+#define ADC_ON_12BIT	(TSC2003_12BIT | TSC2003_ADC_ON_IRQ_DIS0)
+
+#define READ_Y		(ADC_ON_12BIT | TSC2003_MEASURE_Y)
+#define READ_Z1		(ADC_ON_12BIT | TSC2003_MEASURE_Z1)
+#define READ_Z2		(ADC_ON_12BIT | TSC2003_MEASURE_Z2)
+#define READ_X		(ADC_ON_12BIT | TSC2003_MEASURE_X)
+#define PWRDOWN		(TSC2003_12BIT | TSC2003_POWER_OFF_IRQ_EN)
+
+struct ts_event {
+	int	x;
+	int	y;
+	int	z1, z2;
+};
+
+struct tsc2003 {
+	struct input_dev	*input;
+	char			phys[32];
+	struct task_struct	*task;
+	struct ts_event		tc;
+	struct completion 	penirq_completion;
+
+	struct i2c_client	*client;
+
+	u16			model;
+	u16			x_plate_ohms;
+
+	unsigned		pendown;
+};
+
+static inline int tsc2003_xfer(struct tsc2003 *tsc, u8 cmd)
+{
+	s32 data;
+	u16 val;
+
+	data = i2c_smbus_read_word_data(tsc->client, cmd);
+	if (data < 0) {
+		dev_err(&tsc->client->dev, "i2c io error: %d\n", data);
+		return data;
+	}
+
+	/* The protocol and raw data format from i2c interface:
+	 * S Addr Wr [A] Comm [A] S Addr Rd [A] [DataLow] A [DataHigh] NA P
+	 * Where DataLow has [D11-D4], DataHigh has [D3-D0 << 4 | Dummy 4bit].
+	 */
+	val = swab16(data) >> 4;
+
+	dev_dbg(&tsc->client->dev, "data: 0x%x, val: 0x%x\n", data, val);
+
+	return val;
+}
+
+static void tsc2003_send_event(void *tsc)
+{
+	struct tsc2003	*ts = tsc;
+	struct input_dev *input = ts->input;
+	u32		rt = 0;
+	u16		x, y, z1, z2;
+
+	x = ts->tc.x;
+	y = ts->tc.y;
+	z1 = ts->tc.z1;
+	z2 = ts->tc.z2;
+
+	/* range filtering */
+	if (x == MAX_12BIT)
+		x = 0;
+
+	if (likely(x && z1)) {
+		/* compute touch pressure resistance using equation #1 */
+		rt = z2;
+		rt -= z1;
+		rt *= x;
+		rt *= ts->x_plate_ohms;
+		rt /= z1;
+		rt = (rt + 2047) >> 12;
+	}
+
+	/* Sample found inconsistent by debouncing or pressure is beyond
+	 * the maximum. Don't report it to user space, repeat at least
+	 * once more the measurement
+	 */
+	if (rt > MAX_12BIT)
+		return;
+
+	/* NOTE: We can't rely on the pressure to determine the pen down
+	 * state, even this controller has a pressure sensor.  The pressure
+	 * value can fluctuate for quite a while after lifting the pen and
+	 * in some cases may not even settle at the expected value.
+	 *
+	 * The only safe way to check for the pen up condition is in the
+	 * timer by reading the pen signal state (it's a GPIO _and_ IRQ).
+	 */
+	if (rt) {
+		if (!ts->pendown) {
+			dev_dbg(&ts->client->dev, "DOWN\n");
+
+			input_report_key(input, BTN_TOUCH, 1);
+			ts->pendown = 1;
+		}
+
+		input_report_abs(input, ABS_X, x);
+		input_report_abs(input, ABS_Y, y);
+		input_report_abs(input, ABS_PRESSURE, rt);
+
+		input_sync(input);
+
+		dev_dbg(&ts->client->dev, "point(%4d,%4d), pressure (%4u)\n",
+			x, y, rt);
+	} else if (ts->pendown) {
+		/* pen up */
+		dev_dbg(&ts->client->dev, "UP\n");
+		input_report_key(input, BTN_TOUCH, 0);
+		input_report_abs(input, ABS_PRESSURE, 0);
+		input_sync(input);
+
+		ts->pendown = 0;
+	}
+}
+
+static int tsc2003_power_off_irq_en(struct tsc2003 *tsc)
+{
+	/* power down */
+	return tsc2003_xfer(tsc, PWRDOWN);
+}
+
+static int tsc2003_read_values(struct tsc2003 *tsc)
+{
+	/* y- still on; turn on only y+ (and ADC) */
+	tsc->tc.y = tsc2003_xfer(tsc, READ_Y);
+	if (tsc->tc.y < 0)
+		return tsc->tc.y;
+
+	/* turn y- off, x+ on, then leave in lowpower */
+	tsc->tc.x = tsc2003_xfer(tsc, READ_X);
+	if (tsc->tc.x < 0)
+		return tsc->tc.x;
+
+	/* turn y+ off, x- on; we'll use formula #1 */
+	tsc->tc.z1 = tsc2003_xfer(tsc, READ_Z1);
+	if (tsc->tc.z1 < 0)
+		return tsc->tc.z1;
+
+	tsc->tc.z2 = tsc2003_xfer(tsc, READ_Z2);
+	if (tsc->tc.z2 < 0)
+		return tsc->tc.z2;
+
+	return 0;
+}
+
+
+static irqreturn_t tsc2003_irq(int irq, void *handle)
+{
+	struct tsc2003 *ts = handle;
+
+	/* do not call the synced version -> deadlock */
+	disable_irq_nosync(irq);
+	/* signal the thread to continue */
+	complete(&ts->penirq_completion);
+
+	return IRQ_HANDLED;
+}
+
+static int tsc2003_thread(void *d)
+{
+	struct tsc2003 *ts = (struct tsc2003 *)d;
+	int ret;
+
+	allow_signal(SIGKILL);
+
+	while (!signal_pending(current)) {
+		/* power down and wait for interrupt */
+		do {
+			/* loop because the I2C bus might be busy */
+			ret = msleep_interruptible(TS_POLL_PERIOD);
+			if (!ret)
+				ret = tsc2003_power_off_irq_en(ts);
+		} while (ret == -EAGAIN && !signal_pending(current));
+
+		if (signal_pending(current))
+			break;
+
+		ret = wait_for_completion_interruptible(&ts->penirq_completion);
+		if (!ret) {
+			int first = 1;
+			/* got IRQ, start poll, until pen is up */
+			while (!ret && !signal_pending(current)
+				&& (first || ts->pendown)) {
+				ret = tsc2003_read_values(ts);
+				if (!ret)
+					tsc2003_send_event(ts);
+				ret = msleep_interruptible(TS_POLL_PERIOD);
+				first = 0;
+			}
+
+			/* we re enable the interrupt */
+			if (!signal_pending(current))
+				enable_irq(ts->client->irq);
+		}
+	}
+
+	return 0;
+}
+
+static int tsc2003_probe(struct i2c_client *client,
+			const struct i2c_device_id *id)
+{
+	struct tsc2003 *ts;
+	struct tsc2007_platform_data *pdata = client->dev.platform_data;
+	struct input_dev *input_dev;
+	int err;
+
+	if (!pdata) {
+		dev_err(&client->dev, "platform data is required!\n");
+		return -EINVAL;
+	}
+
+	if (!i2c_check_functionality(client->adapter,
+				     I2C_FUNC_SMBUS_READ_WORD_DATA))
+		return -EIO;
+
+	ts = kzalloc(sizeof(struct tsc2003), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!ts || !input_dev) {
+		err = -ENOMEM;
+		goto err_free_mem;
+	}
+
+	ts->client = client;
+	i2c_set_clientdata(client, ts);
+
+	ts->input = input_dev;
+
+	ts->model             = pdata->model;
+	ts->x_plate_ohms      = pdata->x_plate_ohms;
+
+	snprintf(ts->phys, sizeof(ts->phys),
+		 "%s/input0", dev_name(&client->dev));
+
+	input_dev->name = TSC2003_DRIVER_NAME" Touchscreen";
+	input_dev->phys = ts->phys;
+	input_dev->id.bustype = BUS_I2C;
+
+	input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS);
+	input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH);
+
+	input_set_abs_params(input_dev, ABS_X, 0, MAX_12BIT, 0, 0);
+	input_set_abs_params(input_dev, ABS_Y, 0, MAX_12BIT, 0, 0);
+	input_set_abs_params(input_dev, ABS_PRESSURE, 0, MAX_12BIT, 0, 0);
+
+	init_completion(&ts->penirq_completion);
+
+	ts->task = kthread_run(tsc2003_thread, ts, TSC2003_DRIVER_NAME);
+	if (IS_ERR(ts->task)) {
+		err = PTR_ERR(ts->task);
+		goto err_free_mem;
+	}
+
+	err = request_irq(client->irq, tsc2003_irq, 0,
+			client->dev.driver->name, ts);
+	if (err < 0) {
+		dev_err(&client->dev, "irq %d busy?\n", client->irq);
+		goto err_free_thread;
+	}
+
+	err = input_register_device(input_dev);
+	if (err)
+		goto err_free_irq;
+
+	dev_info(&client->dev, "registered with irq (%d)\n", client->irq);
+
+	return 0;
+
+ err_free_irq:
+	free_irq(client->irq, ts);
+ err_free_thread:
+	kthread_stop(ts->task);
+ err_free_mem:
+	input_free_device(input_dev);
+	kfree(ts);
+	return err;
+}
+
+static int tsc2003_remove(struct i2c_client *client)
+{
+	struct tsc2003 *ts = i2c_get_clientdata(client);
+
+	free_irq(client->irq, ts);
+	send_sig(SIGKILL, ts->task, 1);
+	kthread_stop(ts->task);
+	input_unregister_device(ts->input);
+	kfree(ts);
+
+	return 0;
+}
+
+static struct i2c_device_id tsc2003_idtable[] = {
+	{ TSC2003_DRIVER_NAME, 0 },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(i2c, tsc2003_idtable);
+
+static struct i2c_driver tsc2003_driver = {
+	.driver = {
+		.owner	= THIS_MODULE,
+		.name	= TSC2003_DRIVER_NAME,
+		.bus = &i2c_bus_type,
+	},
+	.id_table	= tsc2003_idtable,
+	.probe		= tsc2003_probe,
+	.remove		= tsc2003_remove,
+};
+
+static int __init tsc2003_init(void)
+{
+	return i2c_add_driver(&tsc2003_driver);
+}
+
+static void __exit tsc2003_exit(void)
+{
+	i2c_del_driver(&tsc2003_driver);
+}
+
+module_init(tsc2003_init);
+module_exit(tsc2003_exit);
+
+MODULE_AUTHOR("Mocean Laboratories <[email protected]>");
+MODULE_DESCRIPTION("TSC2003 TouchScreen Driver");
+MODULE_LICENSE("GPL v2");
+
drivers/input/touchscreen/Kconfig +11 -0
Index: linux-2.6.30-rc7/drivers/input/touchscreen/Kconfig
===================================================================
@@ -455,6 +455,17 @@
 	  To compile this driver as a module, choose M here: the
 	  module will be called touchit213.
 
+config TOUCHSCREEN_TSC2003
+	tristate "TSC2003 based touchscreens"
+	depends on I2C
+	help
+	  Say Y here if you have a TSC2003 based touchscreen.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called tsc2003.
+
 config TOUCHSCREEN_TSC2007
 	tristate "TSC2007 based touchscreens"
 	depends on I2C
drivers/input/touchscreen/Makefile +1 -0
Index: linux-2.6.30-rc7/drivers/input/touchscreen/Makefile
===================================================================
@@ -27,6 +27,7 @@
 obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213)	+= touchit213.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT)	+= touchright.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN)	+= touchwin.o
+obj-$(CONFIG_TOUCHSCREEN_TSC2003)	+= tsc2003.o
 obj-$(CONFIG_TOUCHSCREEN_TSC2007)	+= tsc2007.o
 obj-$(CONFIG_TOUCHSCREEN_UCB1400)	+= ucb1400_ts.o
 obj-$(CONFIG_TOUCHSCREEN_WACOM_W8001)	+= wacom_w8001.o
#1 lore raw

Hi,

Quoted 6 lines · Richard Röjfors

Supplied is a driver for the TSC2003. There is actually a driver for TSC2007 which can be used in some cases. The platform struct is reused from the TSC2007.
There is a big difference in the implementation between the drivers, this one does not use HR timers. The TSC2007 driver performs synchronous I2C in the timer callback (SW IRQ context) which is bad when the I2C driver is interrupt driven.

Meaning? I think I2C transaction can sleep. I don't see (right now) need for two drivers. TSC2003 id was added to TSC2007. If you think TSC2007 doesn't provide enough support for it, then please add that in TSC2007 driver itself. We don't need create another driver it seems.

I have added Thierry who can further comment on it, as he has access to device with TSC2003.

Quoted 435 lines

---
Index: linux-2.6.30-rc7/drivers/input/touchscreen/tsc2003.c
===================================================================

drivers/input/touchscreen/tsc2003.c        (revision 869) renamed from drivers/input/touchscreen/tsc2003.c        (revision 0) +387 -0
@@ -0,0 +1,387 @@
+/*
+ * tsc2003.c Driver for TI TSC2003 touch screen controller
+ * Copyright (c) 2009 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/* Supports:
+ * TI TSC2003
+ *
+ * Inspired by tsc2007, Copyright (c) 2008 MtekVision Co., Ltd.
+ */
+#include <linux/module.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/i2c.h>
+#include <linux/i2c/tsc2007.h>
+#include <linux/kthread.h>
+#include <linux/semaphore.h>
+
+#define TSC2003_DRIVER_NAME            "tsc2003"
+
+#define TS_POLL_PERIOD 20              /* ms delay between samples */
+
+#define TSC2003_MEASURE_TEMP0          (0x0 << 4)
+#define TSC2003_MEASURE_AUX            (0x2 << 4)
+#define TSC2003_MEASURE_TEMP1          (0x4 << 4)
+#define TSC2003_ACTIVATE_XN            (0x8 << 4)
+#define TSC2003_ACTIVATE_YN            (0x9 << 4)
+#define TSC2003_ACTIVATE_YP_XN         (0xa << 4)
+#define TSC2003_SETUP                  (0xb << 4)
+#define TSC2003_MEASURE_X              (0xc << 4)
+#define TSC2003_MEASURE_Y              (0xd << 4)
+#define TSC2003_MEASURE_Z1             (0xe << 4)
+#define TSC2003_MEASURE_Z2             (0xf << 4)
+
+#define TSC2003_POWER_OFF_IRQ_EN       (0x0 << 2)
+#define TSC2003_ADC_ON_IRQ_DIS0                (0x1 << 2)
+#define TSC2003_ADC_OFF_IRQ_EN         (0x2 << 2)
+#define TSC2003_ADC_ON_IRQ_DIS1                (0x3 << 2)
+
+#define TSC2003_12BIT                  (0x0 << 1)
+#define TSC2003_8BIT                   (0x1 << 1)
+
+#define        MAX_12BIT                       ((1 << 12) - 1)
+
+#define ADC_ON_12BIT   (TSC2003_12BIT | TSC2003_ADC_ON_IRQ_DIS0)
+
+#define READ_Y         (ADC_ON_12BIT | TSC2003_MEASURE_Y)
+#define READ_Z1                (ADC_ON_12BIT | TSC2003_MEASURE_Z1)
+#define READ_Z2                (ADC_ON_12BIT | TSC2003_MEASURE_Z2)
+#define READ_X         (ADC_ON_12BIT | TSC2003_MEASURE_X)
+#define PWRDOWN                (TSC2003_12BIT | TSC2003_POWER_OFF_IRQ_EN)
+
+struct ts_event {
+       int     x;
+       int     y;
+       int     z1, z2;
+};
+
+struct tsc2003 {
+       struct input_dev        *input;
+       char                    phys[32];
+       struct task_struct      *task;
+       struct ts_event         tc;
+       struct completion       penirq_completion;
+
+       struct i2c_client       *client;
+
+       u16                     model;
+       u16                     x_plate_ohms;
+
+       unsigned                pendown;
+};
+
+static inline int tsc2003_xfer(struct tsc2003 *tsc, u8 cmd)
+{
+       s32 data;
+       u16 val;
+
+       data = i2c_smbus_read_word_data(tsc->client, cmd);
+       if (data < 0) {
+               dev_err(&tsc->client->dev, "i2c io error: %d\n", data);
+               return data;
+       }
+
+       /* The protocol and raw data format from i2c interface:
+        * S Addr Wr [A] Comm [A] S Addr Rd [A] [DataLow] A [DataHigh] NA P
+        * Where DataLow has [D11-D4], DataHigh has [D3-D0 << 4 | Dummy 4bit].
+        */
+       val = swab16(data) >> 4;
+
+       dev_dbg(&tsc->client->dev, "data: 0x%x, val: 0x%x\n", data, val);
+
+       return val;
+}
+
+static void tsc2003_send_event(void *tsc)
+{
+       struct tsc2003  *ts = tsc;
+       struct input_dev *input = ts->input;
+       u32             rt = 0;
+       u16             x, y, z1, z2;
+
+       x = ts->tc.x;
+       y = ts->tc.y;
+       z1 = ts->tc.z1;
+       z2 = ts->tc.z2;
+
+       /* range filtering */
+       if (x == MAX_12BIT)
+               x = 0;
+
+       if (likely(x && z1)) {
+               /* compute touch pressure resistance using equation #1 */
+               rt = z2;
+               rt -= z1;
+               rt *= x;
+               rt *= ts->x_plate_ohms;
+               rt /= z1;
+               rt = (rt + 2047) >> 12;
+       }
+
+       /* Sample found inconsistent by debouncing or pressure is beyond
+        * the maximum. Don't report it to user space, repeat at least
+        * once more the measurement
+        */
+       if (rt > MAX_12BIT)
+               return;
+
+       /* NOTE: We can't rely on the pressure to determine the pen down
+        * state, even this controller has a pressure sensor.  The pressure
+        * value can fluctuate for quite a while after lifting the pen and
+        * in some cases may not even settle at the expected value.
+        *
+        * The only safe way to check for the pen up condition is in the
+        * timer by reading the pen signal state (it's a GPIO _and_ IRQ).
+        */
+       if (rt) {
+               if (!ts->pendown) {
+                       dev_dbg(&ts->client->dev, "DOWN\n");
+
+                       input_report_key(input, BTN_TOUCH, 1);
+                       ts->pendown = 1;
+               }
+
+               input_report_abs(input, ABS_X, x);
+               input_report_abs(input, ABS_Y, y);
+               input_report_abs(input, ABS_PRESSURE, rt);
+
+               input_sync(input);
+
+               dev_dbg(&ts->client->dev, "point(%4d,%4d), pressure (%4u)\n",
+                       x, y, rt);
+       } else if (ts->pendown) {
+               /* pen up */
+               dev_dbg(&ts->client->dev, "UP\n");
+               input_report_key(input, BTN_TOUCH, 0);
+               input_report_abs(input, ABS_PRESSURE, 0);
+               input_sync(input);
+
+               ts->pendown = 0;
+       }
+}
+
+static int tsc2003_power_off_irq_en(struct tsc2003 *tsc)
+{
+       /* power down */
+       return tsc2003_xfer(tsc, PWRDOWN);
+}
+
+static int tsc2003_read_values(struct tsc2003 *tsc)
+{
+       /* y- still on; turn on only y+ (and ADC) */
+       tsc->tc.y = tsc2003_xfer(tsc, READ_Y);
+       if (tsc->tc.y < 0)
+               return tsc->tc.y;
+
+       /* turn y- off, x+ on, then leave in lowpower */
+       tsc->tc.x = tsc2003_xfer(tsc, READ_X);
+       if (tsc->tc.x < 0)
+               return tsc->tc.x;
+
+       /* turn y+ off, x- on; we'll use formula #1 */
+       tsc->tc.z1 = tsc2003_xfer(tsc, READ_Z1);
+       if (tsc->tc.z1 < 0)
+               return tsc->tc.z1;
+
+       tsc->tc.z2 = tsc2003_xfer(tsc, READ_Z2);
+       if (tsc->tc.z2 < 0)
+               return tsc->tc.z2;
+
+       return 0;
+}
+
+
+static irqreturn_t tsc2003_irq(int irq, void *handle)
+{
+       struct tsc2003 *ts = handle;
+
+       /* do not call the synced version -> deadlock */
+       disable_irq_nosync(irq);
+       /* signal the thread to continue */
+       complete(&ts->penirq_completion);
+
+       return IRQ_HANDLED;
+}
+
+static int tsc2003_thread(void *d)
+{
+       struct tsc2003 *ts = (struct tsc2003 *)d;
+       int ret;
+
+       allow_signal(SIGKILL);
+
+       while (!signal_pending(current)) {
+               /* power down and wait for interrupt */
+               do {
+                       /* loop because the I2C bus might be busy */
+                       ret = msleep_interruptible(TS_POLL_PERIOD);
+                       if (!ret)
+                               ret = tsc2003_power_off_irq_en(ts);
+               } while (ret == -EAGAIN && !signal_pending(current));
+
+               if (signal_pending(current))
+                       break;
+
+               ret = wait_for_completion_interruptible(&ts->penirq_completion);
+               if (!ret) {
+                       int first = 1;
+                       /* got IRQ, start poll, until pen is up */
+                       while (!ret && !signal_pending(current)
+                               && (first || ts->pendown)) {
+                               ret = tsc2003_read_values(ts);
+                               if (!ret)
+                                       tsc2003_send_event(ts);
+                               ret = msleep_interruptible(TS_POLL_PERIOD);
+                               first = 0;
+                       }
+
+                       /* we re enable the interrupt */
+                       if (!signal_pending(current))
+                               enable_irq(ts->client->irq);
+               }
+       }
+
+       return 0;
+}
+
+static int tsc2003_probe(struct i2c_client *client,
+                       const struct i2c_device_id *id)
+{
+       struct tsc2003 *ts;
+       struct tsc2007_platform_data *pdata = client->dev.platform_data;
+       struct input_dev *input_dev;
+       int err;
+
+       if (!pdata) {
+               dev_err(&client->dev, "platform data is required!\n");
+               return -EINVAL;
+       }
+
+       if (!i2c_check_functionality(client->adapter,
+                                    I2C_FUNC_SMBUS_READ_WORD_DATA))
+               return -EIO;
+
+       ts = kzalloc(sizeof(struct tsc2003), GFP_KERNEL);
+       input_dev = input_allocate_device();
+       if (!ts || !input_dev) {
+               err = -ENOMEM;
+               goto err_free_mem;
+       }
+
+       ts->client = client;
+       i2c_set_clientdata(client, ts);
+
+       ts->input = input_dev;
+
+       ts->model             = pdata->model;
+       ts->x_plate_ohms      = pdata->x_plate_ohms;
+
+       snprintf(ts->phys, sizeof(ts->phys),
+                "%s/input0", dev_name(&client->dev));
+
+       input_dev->name = TSC2003_DRIVER_NAME" Touchscreen";
+       input_dev->phys = ts->phys;
+       input_dev->id.bustype = BUS_I2C;
+
+       input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS);
+       input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH);
+
+       input_set_abs_params(input_dev, ABS_X, 0, MAX_12BIT, 0, 0);
+       input_set_abs_params(input_dev, ABS_Y, 0, MAX_12BIT, 0, 0);
+       input_set_abs_params(input_dev, ABS_PRESSURE, 0, MAX_12BIT, 0, 0);
+
+       init_completion(&ts->penirq_completion);
+
+       ts->task = kthread_run(tsc2003_thread, ts, TSC2003_DRIVER_NAME);
+       if (IS_ERR(ts->task)) {
+               err = PTR_ERR(ts->task);
+               goto err_free_mem;
+       }
+
+       err = request_irq(client->irq, tsc2003_irq, 0,
+                       client->dev.driver->name, ts);
+       if (err < 0) {
+               dev_err(&client->dev, "irq %d busy?\n", client->irq);
+               goto err_free_thread;
+       }
+
+       err = input_register_device(input_dev);
+       if (err)
+               goto err_free_irq;
+
+       dev_info(&client->dev, "registered with irq (%d)\n", client->irq);
+
+       return 0;
+
+ err_free_irq:
+       free_irq(client->irq, ts);
+ err_free_thread:
+       kthread_stop(ts->task);
+ err_free_mem:
+       input_free_device(input_dev);
+       kfree(ts);
+       return err;
+}
+
+static int tsc2003_remove(struct i2c_client *client)
+{
+       struct tsc2003 *ts = i2c_get_clientdata(client);
+
+       free_irq(client->irq, ts);
+       send_sig(SIGKILL, ts->task, 1);
+       kthread_stop(ts->task);
+       input_unregister_device(ts->input);
+       kfree(ts);
+
+       return 0;
+}
+
+static struct i2c_device_id tsc2003_idtable[] = {
+       { TSC2003_DRIVER_NAME, 0 },
+       { }
+};
+
+MODULE_DEVICE_TABLE(i2c, tsc2003_idtable);
+
+static struct i2c_driver tsc2003_driver = {
+       .driver = {
+               .owner  = THIS_MODULE,
+               .name   = TSC2003_DRIVER_NAME,
+               .bus = &i2c_bus_type,
+       },
+       .id_table       = tsc2003_idtable,
+       .probe          = tsc2003_probe,
+       .remove         = tsc2003_remove,
+};
+
+static int __init tsc2003_init(void)
+{
+       return i2c_add_driver(&tsc2003_driver);
+}
+
+static void __exit tsc2003_exit(void)
+{
+       i2c_del_driver(&tsc2003_driver);
+}
+
+module_init(tsc2003_init);
+module_exit(tsc2003_exit);
+
+MODULE_AUTHOR("Mocean Laboratories <[email protected]>");
+MODULE_DESCRIPTION("TSC2003 TouchScreen Driver");
+MODULE_LICENSE("GPL v2");
+

Index: linux-2.6.30-rc7/drivers/input/touchscreen/Kconfig
===================================================================

drivers/input/touchscreen/Kconfig  (revision 869) renamed from drivers/input/touchscreen/Kconfig  (revision 861) +0 -0
@@ -455,6 +455,17 @@

         To compile this driver as a module, choose M here: the module will be called touchit213.

drivers/input/touchscreen/Kconfig  (revision 869) renamed from drivers/input/touchscreen/Kconfig  (revision 861) +11 -0
+config TOUCHSCREEN_TSC2003
+       tristate "TSC2003 based touchscreens"
+       depends on I2C
+       help
+         Say Y here if you have a TSC2003 based touchscreen.
+
+         If unsure, say N.
+
+         To compile this driver as a module, choose M here: the
+         module will be called tsc2003.
+

 config TOUCHSCREEN_TSC2007
       tristate "TSC2007 based touchscreens" depends on I2C
Index: linux-2.6.30-rc7/drivers/input/touchscreen/Makefile
===================================================================

drivers/input/touchscreen/Makefile (revision 869) renamed from drivers/input/touchscreen/Makefile (revision 861) +0 -0
@@ -27,6 +27,7 @@

 obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213)   += touchit213.o obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT)   += touchright.o obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN)     += touchwin.o +obj-$(CONFIG_TOUCHSCREEN_TSC2003)      += tsc2003.o obj-$(CONFIG_TOUCHSCREEN_TSC2007)      += tsc2007.o obj-$(CONFIG_TOUCHSCREEN_UCB1400)      += ucb1400_ts.o obj-$(CONFIG_TOUCHSCREEN_WACOM_W8001)  += wacom_w8001.o
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Signature · 7 lines
---Trilok Soni
http://triloksoni.wordpress.com
http://www.linkedin.com/in/triloksoni
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#2 lore raw
Quoted 4 lines · Trilok Soni

Hi,

On Mon, Jun 15, 2009 at 8:39 PM, Richard Röjfors<[email protected]> wrote:

Quoted 8 lines
Quoted 6 lines

Supplied is a driver for the TSC2003. There is actually a driver for TSC2007 which can be used in some cases. The platform struct is reused from the TSC2007.
There is a big difference in the implementation between the drivers, this one does not use HR timers. The TSC2007 driver performs synchronous I2C in the timer callback (SW IRQ context) which is bad when the I2C driver is interrupt driven.

Meaning? I think I2C transaction can sleep.

Yes that's what it means, and that's bad in a HR timer callback.

Quoted 4 lines

I don't see (right now)
need for two drivers. TSC2003 id was added to TSC2007. If you think TSC2007 doesn't provide enough support for it, then please add that in TSC2007 driver itself. We don't need create another driver it seems.

The chips are actually compatible (when it comes to the touchscreen functionality).

So you are true, the tsc2007 can be modified, I didn't added support for the other pins anyway.

Actually I don't have access to any tsc2007, so I can't verify it on the newer chip. I will go on and propose a patch.

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

#3 lore raw
Quoted 12 lines · Richard Röjfors
Quoted 11 lines · Trilok Soni

Hi,

Quoted 6 lines · Richard Röjfors

Supplied is a driver for the TSC2003. There is actually a driver for TSC2007 which can be used in some cases. The platform struct is reused from the TSC2007.
There is a big difference in the implementation between the drivers, this one does not use HR timers. The TSC2007 driver performs synchronous I2C in the timer callback (SW IRQ context) which is bad when the I2C driver is interrupt driven.

Quoted 3 lines
Quoted 1 line

Meaning? I think I2C transaction can sleep.

Yes that's what it means, and that's bad in a HR timer callback.

Note that my patch to the tsc2007 to support the tsc2003 exactly fixes this problem. It moves the actual I2C transfers into a workqueue, so no sleeping functions are called from the hrtimer callback.

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

#4 lore raw
Quoted 7 lines · Thierry Reding
Quoted 3 lines
Quoted 2 lines
Quoted 1 line

Meaning? I think I2C transaction can sleep.

Yes that's what it means, and that's bad in a HR timer callback.

Note that my patch to the tsc2007 to support the tsc2003 exactly fixes this problem. It moves the actual I2C transfers into a workqueue, so no sleeping functions are called from the hrtimer callback.

Ah good! The IRQ is disabled no sync too.
We actually don't have the possibility to implement a pen down state callback, so I need to modify the code slightly to work even without one. (but not be as accurate when the callback is not available)

Your patch where it schedules work rather than calling the I2C function directly isn't in mainline.
I saw a patch where you added the work scheduling, and a later patch where you fixed some spinlock stuff, have you resent the patch for the work scheduling?

--Richard

#5 lore raw

[Cc'ing Kwangwoo Lee]

Quoted 14 lines · Richard Röjfors
Quoted 7 lines · Thierry Reding
Quoted 3 lines
Quoted 2 lines
Quoted 1 line

Meaning? I think I2C transaction can sleep.

Yes that's what it means, and that's bad in a HR timer callback.

Note that my patch to the tsc2007 to support the tsc2003 exactly fixes this problem. It moves the actual I2C transfers into a workqueue, so no sleeping functions are called from the hrtimer callback.

Ah good! The IRQ is disabled no sync too.
We actually don't have the possibility to implement a pen down state callback, so I need to modify the code slightly to work even without one. (but not be as accurate when the callback is not available)

Quoted 4 lines

Your patch where it schedules work rather than calling the I2C function directly isn't in mainline.
I saw a patch where you added the work scheduling, and a later patch where you fixed some spinlock stuff, have you resent the patch for the work scheduling?

I think there were still some issues with that patch. Kwangwoo Lee was the last to comment. This is from the previous thread:

Quoted 26 lines
Quoted 12 lines · Dmitry Torokhov
Quoted 11 lines · Dmitry Torokhov

Hi,

Quoted 9 lines · Kwangwoo Lee

From d5de0d22109de7564f9bf1df688acbe6b18f41db Mon Sep 17 00:00:00 2001
From: Kwangwoo Lee <[email protected]>
Date: Mon, 11 May 2009 20:05:50 +0900
Subject: [PATCH 2/2] Input: tsc2007: do I2C transfers in non-interrupt context.

This patch enhances pointer movements much smoother. The original patch is written by Thierry.

Quoted 11 lines
Quoted 11 lines

drivers/input/touchscreen/tsc2007.c struct ts_event { · line 73

Quoted 7 lines
drivers/input/touchscreen/tsc2007.c +1 -0
@@ -70,6 +70,7 @@ struct ts_event {
  struct tsc2007 {
     struct input_dev        *input;
     char                    phys[32];
+   struct work_struct      work;

Every time I see a work_struct in a driver and don't see cancel_work_sync() anywhere I know there are issues...

Quoted 7 lines

Thanks for your comment. I also missed that thing, sorry.

Quoted 2 lines

Also, why do we need to chain irq->timer->work now? Surely we can bypass the timer if we have to read in process context.

It's good point. I'll check again.
Thanks.

So perhaps some more work is required to get this mainlined.

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

#6 lore raw

Hi,

Quoted 53 lines · Thierry Reding

[Cc'ing Kwangwoo Lee]

Quoted 14 lines · Richard Röjfors
Quoted 7 lines · Thierry Reding
Quoted 3 lines
Quoted 2 lines
Quoted 1 line

Meaning? I think I2C transaction can sleep.

Yes that's what it means, and that's bad in a HR timer callback.

Note that my patch to the tsc2007 to support the tsc2003 exactly fixes this problem. It moves the actual I2C transfers into a workqueue, so no sleeping functions are called from the hrtimer callback.

Ah good! The IRQ is disabled no sync too.
We actually don't have the possibility to implement a pen down state callback, so I need to modify the code slightly to work even without one. (but not be as accurate when the callback is not available)

Quoted 4 lines

Your patch where it schedules work rather than calling the I2C function directly isn't in mainline.
I saw a patch where you added the work scheduling, and a later patch where you fixed some spinlock stuff, have you resent the patch for the work scheduling?

I think there were still some issues with that patch. Kwangwoo Lee was the last to comment. This is from the previous thread:

Quoted 28 lines
Quoted 12 lines · Dmitry Torokhov
Quoted 11 lines · Dmitry Torokhov

Hi,

Quoted 9 lines · Kwangwoo Lee

From d5de0d22109de7564f9bf1df688acbe6b18f41db Mon Sep 17 00:00:00 2001
From: Kwangwoo Lee <[email protected]>
Date: Mon, 11 May 2009 20:05:50 +0900
Subject: [PATCH 2/2] Input: tsc2007: do I2C transfers in non-interrupt context.

This patch enhances pointer movements much smoother. The original patch is written by Thierry.

Quoted 11 lines
Quoted 11 lines

drivers/input/touchscreen/tsc2007.c struct ts_event { · line 73

Quoted 7 lines
drivers/input/touchscreen/tsc2007.c +1 -0
@@ -70,6 +70,7 @@ struct ts_event {
  struct tsc2007 {
     struct input_dev        *input;
     char                    phys[32];
+   struct work_struct      work;

Every time I see a work_struct in a driver and don't see cancel_work_sync() anywhere I know there are issues...

Thanks for your comment. I also missed that thing, sorry.

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

Also, why do we need to chain irq->timer->work now? Surely we can bypass the timer if we have to read in process context.

It's good point. I'll check again.
Thanks.

So perhaps some more work is required to get this mainlined.

Sorry for my late reply.

I think that the driver need to be fixed to send i2c command asynchronously (if possible)
or replace hrtimer with delayed workqueue as Dmitry pointed out above.

Hmm.. I'm going to move to the other company, so I do not have the testing environment anymore.
I can see the emails from the list, but can anyone else follow up the driver, please?

Thanks.

Signature · 5 lines
Kwangwoo Lee <[email protected]>
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html