From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
To: Qipeng Zha <qipeng.zha@intel.com>
Cc: linux-input@vger.kernel.org
Subject: Re: [PATCH] input:keyboard: add Cypress button driver for Intel Platform
Date: Fri, 9 Oct 2015 17:40:48 -0700 [thread overview]
Message-ID: <20151010004048.GA14662@dtor-ws> (raw)
In-Reply-To: <1444385976-57339-1-git-send-email-qipeng.zha@intel.com>
Hi Qipeng,
On Fri, Oct 09, 2015 at 06:19:36PM +0800, Qipeng Zha wrote:
> This driver is to support Cypress CY8CMBR3XXX family controller,
> which is used as button input for Intel platforms.
Overall it does not look like pure keyboard device, I'd move into
drivers/input/misc. Are you planning on supporting sliteds, leds, etc?
Also, why only Intel platforms?
>
> Signed-off-by: Qipeng Zha <qipeng.zha@intel.com>
> ---
> drivers/input/keyboard/Kconfig | 7 +
> drivers/input/keyboard/Makefile | 1 +
> drivers/input/keyboard/cypress-keyboard.c | 278 ++++++++++++++++++++++++++++++
> 3 files changed, 286 insertions(+)
> create mode 100644 drivers/input/keyboard/cypress-keyboard.c
>
> diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
> index a89ba7c..f39f148 100644
> --- a/drivers/input/keyboard/Kconfig
> +++ b/drivers/input/keyboard/Kconfig
> @@ -686,4 +686,11 @@ config KEYBOARD_CAP11XX
> To compile this driver as a module, choose M here: the
> module will be called cap11xx.
>
> +config KEYBOARD_CYPRESS_BUTTON
> + tristate "Cypress Button"
Please be a bit more descriptive.
> + depends on I2C
> + help
> + Say Y here to enable support for Cypress button for
> + Intel platforms.
To compile this dirver as module...
> endif
> diff --git a/drivers/input/keyboard/Makefile b/drivers/input/keyboard/Makefile
> index 4707678..5a73138 100644
> --- a/drivers/input/keyboard/Makefile
> +++ b/drivers/input/keyboard/Makefile
> @@ -60,3 +60,4 @@ obj-$(CONFIG_KEYBOARD_TEGRA) += tegra-kbc.o
> obj-$(CONFIG_KEYBOARD_TWL4030) += twl4030_keypad.o
> obj-$(CONFIG_KEYBOARD_XTKBD) += xtkbd.o
> obj-$(CONFIG_KEYBOARD_W90P910) += w90p910_keypad.o
> +obj-$(CONFIG_KEYBOARD_CYPRESS_BUTTON) += cypress-keyboard.o
> diff --git a/drivers/input/keyboard/cypress-keyboard.c b/drivers/input/keyboard/cypress-keyboard.c
> new file mode 100644
> index 0000000..4a2e7f7
> --- /dev/null
> +++ b/drivers/input/keyboard/cypress-keyboard.c
> @@ -0,0 +1,278 @@
> +/*
> + * Cypress button driver
> + *
> + * Copyright (c) 2015, Intel Corporation.
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms of the GNU General Public License as published by the
> + * Free Software Foundation; either version 2 of the License, or (at your
> + * option) any later version.
> + */
> +
> +#include <linux/module.h>
> +#include <linux/init.h>
> +#include <linux/delay.h>
> +#include <linux/gpio.h>
> +#include <linux/gpio/consumer.h>
> +#include <linux/i2c.h>
> +#include <linux/interrupt.h>
> +#include <linux/irq.h>
> +#include <linux/slab.h>
> +#include <linux/input.h>
> +#include <linux/acpi.h>
> +
> +/* registers */
> +#define SENSOR_EN 0x0
> +#define SENSOR_CS0 0
> +#define SENSOR_CS1 1
> +#define BUTTON_STAT 0xAA
> +#define BUTTON_ACTIVE 1
> +
> +/* retry I2C transfer in case of autosleep */
> +#define I2C_RETRY_TIMES 10
> +
> +struct cy_kb {
> + struct i2c_client *client;
> + struct input_dev *input_dev;
> + struct mutex mutex;
> + unsigned int irq;
> + int gpio;
> + u16 button_status;
> + bool suspended;
> +};
> +
> +static int cy_kb_read(struct cy_kb *data,
> + u16 reg, u16 len, void *buf)
> +{
> + int ret;
> + u8 regbuf[1];
> + struct i2c_msg xfer[2];
> + int retry = I2C_RETRY_TIMES;
> +
> + regbuf[0] = reg & 0xff;
> +
> + xfer[0].addr = data->client->addr;
> + xfer[0].flags = 0;
> + xfer[0].len = sizeof(regbuf);
> + xfer[0].buf = regbuf;
> +
> + xfer[1].addr = data->client->addr;
> + xfer[1].flags = I2C_M_RD;
> + xfer[1].len = len;
> + xfer[1].buf = buf;
> +
> +retry_read:
> + ret = i2c_transfer(data->client->adapter, xfer, ARRAY_SIZE(xfer));
> + if (ret != ARRAY_SIZE(xfer)) {
> + if (retry--) {
> + dev_dbg(&data->client->dev, "i2c read retry\n");
> + goto retry_read;
> + } else {
> + dev_err(&data->client->dev, "i2c transfer failed (%d)\n",
> + ret);
> + return -EIO;
> + }
> + }
> +
> + return 0;
> +}
> +
> +static int cy_kb_write(struct cy_kb *data, u8 reg, u16 val)
> +{
> + int ret;
> + u8 *buf;
> + int count;
> + int retry = I2C_RETRY_TIMES;
> +
> + count = sizeof(val) + 1;
> + buf = kmalloc(count, GFP_KERNEL);
> + if (!buf)
> + return -ENOMEM;
> +
> + buf[0] = reg & 0xff;
> + memcpy(&buf[1], &val, sizeof(val));
> +
> +retry_write:
> + ret = i2c_master_send(data->client, buf, count);
> + if (ret != count) {
> + if (retry--) {
> + dev_dbg(&data->client->dev, "i2c write retry\n");
> + goto retry_write;
> + } else {
> + dev_err(&data->client->dev, "i2c send failed (%d)\n",
> + ret);
> + ret = -EIO;
> + }
> + } else {
> + ret = 0;
> + }
> +
> + kfree(buf);
> + return ret;
> +}
> +
> +static irqreturn_t cy_kb_interrupt(int irq, void *dev_id)
> +{
> + int ret;
> + u8 buf = 0x0;
> + struct cy_kb *data = dev_id;
> +
> + mutex_lock(&data->mutex);
Why do you need this mutex? What can you be racing with?
> +
> + ret = cy_kb_read(data, BUTTON_STAT, sizeof(buf), &buf);
> + if (ret) {
> + dev_err(&data->client->dev, "can't read button status\n");
> + goto out;
> + }
> + dev_dbg(&data->client->dev, "Button status 0x%02x\n", buf);
> +
> + if ((buf ^ data->button_status) & (BUTTON_ACTIVE << SENSOR_CS0)) {
> + input_event(data->input_dev, EV_KEY, KEY_BACK,
> + (buf >> SENSOR_CS0) & BUTTON_ACTIVE);
> + }
> +
> + if ((buf ^ data->button_status) & (BUTTON_ACTIVE << SENSOR_CS1)) {
> + input_event(data->input_dev, EV_KEY, KEY_MENU,
> + (buf >> SENSOR_CS1) & BUTTON_ACTIVE);
> + }
> +
> + input_sync(data->input_dev);
> + data->button_status = buf;
> +out:
> + mutex_unlock(&data->mutex);
> + return IRQ_HANDLED;
> +}
> +
> +static int cy_kb_probe(struct i2c_client *client,
> + const struct i2c_device_id *id)
> +{
> + int ret;
> + struct cy_kb *data;
> + struct gpio_desc *gpio;
> +
> + data = devm_kzalloc(&client->dev, sizeof(struct cy_kb), GFP_KERNEL);
> + if (!data)
> + return -ENOMEM;
> +
> + i2c_set_clientdata(client, data);
> + mutex_init(&data->mutex);
> + data->client = client;
> + data->irq = client->irq;
> + gpio = devm_gpiod_get_index(&client->dev, "cypress_gpio_int", 0);
> + if (!IS_ERR(gpio)) {
> + data->gpio = desc_to_gpio(gpio);
> + data->irq = gpiod_to_irq(gpio_to_desc(data->gpio));
> + } else
> + dev_err(&client->dev, "Failed to get gpio interrupt\n");
Why do you need gpio if you are not using it as gpio? Platform should
set right irq in client structure.
> +
> + ret = request_threaded_irq(data->irq, NULL, cy_kb_interrupt,
> + IRQF_TRIGGER_LOW | IRQF_ONESHOT,
I'd rather we relied on platform to set up interrupt trigger.
> + client->name, data);
> + if (ret) {
> + dev_err(&client->dev, "Failed to register irq\n");
> + return ret;
> + }
> +
> + data->input_dev = input_allocate_device();
This is too late: IRQ may already arrive before you allocated the
device.
> + if (!data->input_dev) {
> + dev_err(&client->dev, "Failed to allocate input device\n");
> + ret = -ENOMEM;
> + goto err_alloc_input;
> + }
> + data->input_dev->name = client->name;
> + data->input_dev->id.bustype = BUS_I2C;
> + data->input_dev->dev.parent = &data->client->dev;
> + input_set_capability(data->input_dev, EV_KEY, KEY_BACK);
> + input_set_capability(data->input_dev, EV_KEY, KEY_MENU);
Can we pass the button codes via DT/ACPI properties instead of
hard coding?
> + ret = input_register_device(data->input_dev);
> + if (ret) {
> + dev_err(&client->dev, "Failed to register input device\n");
> + goto err_reg_input;
> + }
> +
> + return 0;
> +
> +err_reg_input:
> + input_free_device(data->input_dev);
> +err_alloc_input:
> + free_irq(data->irq, data);
> + gpio_free(data->gpio);
> +
> + return ret;
> +}
> +
> +static int cy_kb_remove(struct i2c_client *client)
> +{
> + struct cy_kb *data = i2c_get_clientdata(client);
> +
> + free_irq(data->irq, data);
> + gpio_free(data->gpio);
> + input_unregister_device(data->input_dev);
> +
> + return 0;
> +}
> +
> +#ifdef CONFIG_PM_SLEEP
> +static int cy_kb_suspend(struct device *dev)
Please use __maybe_unused instead of #ifdef.
> +{
> + struct i2c_client *client = to_i2c_client(dev);
> + struct cy_kb *data = i2c_get_clientdata(client);
> +
> + mutex_lock(&data->mutex);
> + if (!data->suspended)
How can it be suspended here?
> + cy_kb_write(data, SENSOR_EN, 0);
> + data->suspended = true;
> + mutex_unlock(&data->mutex);
> +
> + return 0;
> +}
> +
> +static int cy_kb_resume(struct device *dev)
> +{
> + struct i2c_client *client = to_i2c_client(dev);
> + struct cy_kb *data = i2c_get_clientdata(client);
> +
> + mutex_lock(&data->mutex);
> + if (data->suspended) {
> + u16 val = (1 << SENSOR_CS0) | (1 << SENSOR_CS1);
> +
> + cy_kb_write(data, SENSOR_EN, val);
> + }
> + data->suspended = false;
> + mutex_unlock(&data->mutex);
> +
> + return 0;
> +}
> +static SIMPLE_DEV_PM_OPS(cy_kb_pm_ops, cy_kb_suspend, cy_kb_resume);
> +#endif
> +
> +static const struct i2c_device_id cy_kb_id[] = {
> + { "CY8M3108:00", 0 },
> + { }
> +};
> +MODULE_DEVICE_TABLE(i2c, cy_kb_id);
> +
> +static struct acpi_device_id cy_kb_acpi_id[] = {
> + { "CY8M3108", 0 },
> + { },
> +};
> +MODULE_DEVICE_TABLE(acpi, cy_kb_acpi_id);
> +
> +static struct i2c_driver cy_kb_driver = {
> + .driver = {
> + .name = "cypressbutton",
> + .acpi_match_table = ACPI_PTR(cy_kb_acpi_id),
> +#ifdef CONFIG_PM_SLEEP
> + .pm = &cy_kb_pm_ops,
> +#endif
> + },
> + .probe = cy_kb_probe,
> + .remove = cy_kb_remove,
> + .id_table = cy_kb_id,
> +};
> +
> +module_i2c_driver(cy_kb_driver);
> +
> +MODULE_AUTHOR("Qipeng Zha<qipeng.zha@intel.com>");
> +MODULE_DESCRIPTION("Cypress CY8CMBRXXX driver");
> +MODULE_LICENSE("GPL v2");
The copyright notice at the beginning of the file mentions GPL v2+ but
MODULE_LICENSE states v2 only, please make consistent.
Thanks.
--
Dmitry
next prev parent reply other threads:[~2015-10-10 0:40 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-09 10:19 [PATCH] input:keyboard: add Cypress button driver for Intel Platform Qipeng Zha
2015-10-10 0:40 ` Dmitry Torokhov [this message]
-- strict thread matches above, loose matches on Subject: below --
2015-09-22 14:02 Qipeng Zha
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20151010004048.GA14662@dtor-ws \
--to=dmitry.torokhov@gmail.com \
--cc=linux-input@vger.kernel.org \
--cc=qipeng.zha@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).