Skip to contentSkip to conversation

Input/Touchscreen Driver: add support AD7877 touchscreen driver (resend to linux-input mailist)

PATCH In review Bryan Wu 3 participants 3 messages Feb 4, 2008 – Feb 5, 2008

Subsystems (3)
lore
Share & tools
lore

Bryan Wu

author
#0 lore raw

PATCH 1/1 Input/Touchscreen Driver: add support AD7877 touchscreen driver (resend to linux-input mailist)

Quoted 1 line

From a9a37412a96d693fab23cb2ecdd5d1e86b02fad1 Mon Sep 17 00:00:00 2001

From: Michael Hennerich <[email protected]>
Date: Fri, 25 Jan 2008 15:42:53 +0800
Subject: [PATCH] Input/Touchscreen Driver: add support AD7877 touchscreen driver

[try #2] Changelog:
- move locking inside ad7877_enable and ad7877_disable
- use setup_timer
- use input_dev->dev.parent
- fix unregister device
- kill EV_KEY since it's not used
- fix indention style

Apply patch from Philip Douglass:
Make ad7877 ts driver not dependent on Blackfin specific cs_change_per_word_option

---
 drivers/input/touchscreen/Kconfig  |   13 +
 drivers/input/touchscreen/Makefile |    1 +
 drivers/input/touchscreen/ad7877.c |  929 ++++++++++++++++++++++++++++++++++++
 3 files changed, 943 insertions(+), 0 deletions(-)
 create mode 100644 drivers/input/touchscreen/ad7877.c
drivers/input/touchscreen/Kconfig +13 -0
index 90e8e92..94845df 100644
@@ -29,6 +29,19 @@ config TOUCHSCREEN_ADS7846
 	  To compile this driver as a module, choose M here: the
 	  module will be called ads7846.
 
+config TOUCHSCREEN_AD7877
+	tristate "AD7877 based touchscreens"
+	depends on SPI_MASTER
+	help
+	  Say Y here if you have a touchscreen interface using the
+	  AD7877 controller, and your board-specific initialization
+	  code includes that in its table of SPI devices.
+
+	  If unsure, say N (but it's safe to say "Y").
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called ad7877.
+
 config TOUCHSCREEN_BITSY
 	tristate "Compaq iPAQ H3600 (Bitsy) touchscreen"
 	depends on SA1100_BITSY
drivers/input/touchscreen/Makefile +1 -0
index 35d4097..49561b2 100644
@@ -4,6 +4,7 @@
 
 # Each configuration option enables a list of files.
 
+obj-$(CONFIG_TOUCHSCREEN_AD7877)	+= ad7877.o
 obj-$(CONFIG_TOUCHSCREEN_ADS7846)	+= ads7846.o
 obj-$(CONFIG_TOUCHSCREEN_BITSY)		+= h3600_ts_input.o
 obj-$(CONFIG_TOUCHSCREEN_CORGI)		+= corgi_ts.o
drivers/input/touchscreen/ad7877.c new file +929 -0

929 lines not on this page · raw message

Signature · 1 line
1.5.3.4
maintainer
#1 lore raw

Hi Bryan,

On Mon, Feb 04, 2008 at 05:52:38PM +0800, Bryan Wu wrote:
+

drivers/input/touchscreen/ad7877.c line 251

Quoted 8 lines
drivers/input/touchscreen/ad7877.c new file +8 -0
+	spi_message_add_tail(&req->xfer[0], &req->msg);
+	spi_message_add_tail(&req->xfer[1], &req->msg);
+
+	status = spi_sync(spi, &req->msg);
+
+	if (req->msg.status)
+		status = req->msg.status;
+

I think the "new way" for spi is to just check return value of spi_sync and not bother with req->msg.status, but even using old style spi_sycn I'd expect something like:

	status = spi_sync(spi, &req->msg);
	if (status == 0)
		status = req->msg.status;
Quoted 6 lines
drivers/input/touchscreen/ad7877.c new file +6 -0
+
+static ssize_t ad7877_gpio3_store(struct device *dev,
+				     struct device_attribute *attr,
+				     const char *buf, size_t count)
+{
+	struct ad7877 *ts = dev_get_drvdata(dev);

drivers/input/touchscreen/ad7877.c line 438

Quoted 10 lines
drivers/input/touchscreen/ad7877.c new file +10 -0
+	char *endp;
+	int i;
+
+	i = simple_strtoul(buf, &endp, 10);
+
+	if (i)
+		ts->gpio3=1;
+	else
+		ts->gpio3=0;
+

I am tempted to change the above to ts->gpio3 = !!i;

Quoted 13 lines
drivers/input/touchscreen/ad7877.c new file +13 -0
+
+	if (likely(x && z1 && !device_suspended(&ts->spi->dev))) {
+		/* compute touch pressure resistance using equation #2 */
+		Rt = z2;
+		Rt -= z1;
+		Rt *= x;
+		Rt *= ts->x_plate_ohms;
+		Rt /= z1;
+		Rt = (Rt + 2047) >> 12;
+	} else
+		Rt = 0;
+
+	if (Rt) {

drivers/input/touchscreen/ad7877.c line 543

Quoted 10 lines
drivers/input/touchscreen/ad7877.c new file +10 -0
+		input_report_abs(input_dev, ABS_X, x);
+		input_report_abs(input_dev, ABS_Y, y);
+		input_report_abs(input_dev, ABS_PRESSURE, Rt);
+		input_sync(input_dev);
+	}
+
+#ifdef	VERBOSE
+	if (Rt)
+		pr_debug("%s: %d/%d/%d%s\n", ts->spi->dev.bus_id,
+			x, y, Rt, Rt ? "" : " UP");

We check the same condition 3 times in a row. The compiler might combine them but why not help it?

drivers/input/touchscreen/ad7877.c line 667

Quoted 6 lines
drivers/input/touchscreen/ad7877.c new file +6 -0
+
+static int ad7877_suspend(struct spi_device *spi, pm_message_t message)
+{
+	struct ad7877 *ts = dev_get_drvdata(&spi->dev);
+
+	spi->dev.power.power_state = message;

I think they are trying to get rid of power_state...

drivers/input/touchscreen/ad7877.c line 678

Quoted 11 lines
drivers/input/touchscreen/ad7877.c new file +11 -0
+	ad7877_disable(ts);
+
+	return 0;
+
+}
+
+static int ad7877_resume(struct spi_device *spi)
+{
+	struct ad7877 *ts = dev_get_drvdata(&spi->dev);
+
+	spi->dev.power.power_state = PMSG_ON;

Same here.

drivers/input/touchscreen/ad7877.c line 855

Quoted 11 lines
drivers/input/touchscreen/ad7877.c new file +11 -0
+
+	err = input_register_device(input_dev);
+	if (err)
+		goto err_idev;
+
+	ts->intr_flag = 0;
+
+	return 0;
+
+err_idev:
+	input_dev = NULL; /* so we don't try to free it later */

But we _do_ need to free it if input_register_device() fails.

Thanks.

Signature · 1 line
Dmitry
#2 lore raw

RE: [PATCH] Input/Touchscreen Driver: add support AD7877touchscreen driver (resend to linux-input mailist)

Dmitry,

Thanks for you feedback.
Bryan is going to send out an updated patch.

Best regards,
Michael

Quoted 48 lines

-----Original Message-----
From: Dmitry Torokhov [mailto:[email protected]]
Sent: Montag, 4. Februar 2008 18:13
To: Bryan Wu
Cc: Michael Hennerich; [email protected]
Subject: Re: [PATCH] Input/Touchscreen Driver: add support AD7877touchscreen driver (resend to linux-input mailist)

Hi Bryan,

On Mon, Feb 04, 2008 at 05:52:38PM +0800, Bryan Wu wrote:
+

drivers/input/touchscreen/ad7877.c line 251

Quoted 8 lines
drivers/input/touchscreen/ad7877.c new file +8 -0
+	spi_message_add_tail(&req->xfer[0], &req->msg);
+	spi_message_add_tail(&req->xfer[1], &req->msg);
+
+	status = spi_sync(spi, &req->msg);
+
+	if (req->msg.status)
+		status = req->msg.status;
+

I think the "new way" for spi is to just check return value of spi_sync and not bother with req->msg.status, but even using old style spi_sycn I'd expect something like:

	status = spi_sync(spi, &req->msg);
	if (status == 0)
		status = req->msg.status;
Quoted 6 lines
drivers/input/touchscreen/ad7877.c new file +6 -0
+
+static ssize_t ad7877_gpio3_store(struct device *dev,
+				     struct device_attribute *attr,
+				     const char *buf, size_t count)
+{
+	struct ad7877 *ts = dev_get_drvdata(dev);

drivers/input/touchscreen/ad7877.c line 438

Quoted 10 lines
drivers/input/touchscreen/ad7877.c new file +10 -0
+	char *endp;
+	int i;
+
+	i = simple_strtoul(buf, &endp, 10);
+
+	if (i)
+		ts->gpio3=1;
+	else
+		ts->gpio3=0;
+

I am tempted to change the above to ts->gpio3 = !!i;

Quoted 3 lines

drivers/input/touchscreen/ad7877.c

Quoted 3 lines
drivers/input/touchscreen/ad7877.c new file +3 -0
+
+	if (likely(x && z1 && !device_suspended(&ts->spi->dev))) {
+		/* compute touch pressure resistance using equation #2

*/

Quoted 24 lines
Quoted 10 lines
drivers/input/touchscreen/ad7877.c new file +10 -0
+		Rt = z2;
+		Rt -= z1;
+		Rt *= x;
+		Rt *= ts->x_plate_ohms;
+		Rt /= z1;
+		Rt = (Rt + 2047) >> 12;
+	} else
+		Rt = 0;
+
+	if (Rt) {

drivers/input/touchscreen/ad7877.c line 543

Quoted 10 lines
drivers/input/touchscreen/ad7877.c new file +10 -0
+		input_report_abs(input_dev, ABS_X, x);
+		input_report_abs(input_dev, ABS_Y, y);
+		input_report_abs(input_dev, ABS_PRESSURE, Rt);
+		input_sync(input_dev);
+	}
+
+#ifdef	VERBOSE
+	if (Rt)
+		pr_debug("%s: %d/%d/%d%s\n", ts->spi->dev.bus_id,
+			x, y, Rt, Rt ? "" : " UP");

We check the same condition 3 times in a row. The compiler might combine them but why not help it?

Quoted 2 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+
+static int ad7877_suspend(struct spi_device *spi, pm_message_t

message)

Quoted 39 lines

drivers/input/touchscreen/ad7877.c line 667

Quoted 4 lines
drivers/input/touchscreen/ad7877.c new file +4 -0
+{
+	struct ad7877 *ts = dev_get_drvdata(&spi->dev);
+
+	spi->dev.power.power_state = message;

I think they are trying to get rid of power_state...

drivers/input/touchscreen/ad7877.c line 678

Quoted 11 lines
drivers/input/touchscreen/ad7877.c new file +11 -0
+	ad7877_disable(ts);
+
+	return 0;
+
+}
+
+static int ad7877_resume(struct spi_device *spi)
+{
+	struct ad7877 *ts = dev_get_drvdata(&spi->dev);
+
+	spi->dev.power.power_state = PMSG_ON;

Same here.

drivers/input/touchscreen/ad7877.c line 855

Quoted 11 lines
drivers/input/touchscreen/ad7877.c new file +11 -0
+
+	err = input_register_device(input_dev);
+	if (err)
+		goto err_idev;
+
+	ts->intr_flag = 0;
+
+	return 0;
+
+err_idev:
+	input_dev = NULL; /* so we don't try to free it later */

But we _do_ need to free it if input_register_device() fails.

Thanks.

--
Dmitry