[PATCH 1/3] Input: imx_keypad - Add device tree support
Liu Ying
liu.y.victor at gmail.com
Mon Dec 31 01:36:54 EST 2012
Hi, Dmitry,
Thanks for reviewing this patch.
2012/12/31 Dmitry Torokhov <dmitry.torokhov at gmail.com>:
> Hi Liu,
>
> On Sun, Dec 30, 2012 at 09:09:04PM +0800, Liu Ying wrote:
>> This patch adds device tree support for imx keypad driver.
>> 1) Basic device tree binding support.
>> 2) Since the probe function needs keymap and keymap size
>> information, we get them from device tree if it is
>> supported, otherwise, we fall back to legacy platform data.
>> 3) Support to configure keypad pins via pinctrl system.
>>
>> Signed-off-by: Liu Ying <Ying.Liu at freescale.com>
>> ---
>> .../devicetree/bindings/input/imx-keypad.txt | 54 +++++++++++++
>> drivers/input/keyboard/imx_keypad.c | 80 ++++++++++++++++++--
>> 2 files changed, 127 insertions(+), 7 deletions(-)
>> create mode 100644 Documentation/devicetree/bindings/input/imx-keypad.txt
>>
>> diff --git a/Documentation/devicetree/bindings/input/imx-keypad.txt b/Documentation/devicetree/bindings/input/imx-keypad.txt
>> new file mode 100644
>> index 0000000..1c6e16d
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/input/imx-keypad.txt
>> @@ -0,0 +1,54 @@
>> +* Freescale i.MX Keypad Port(KPP) device tree bindings
>> +
>> +The KPP is designed to interface with a keypad matrix with 2-point contact
>> +or 3-point contact keys. The KPP is designed to simplify the software task
>> +of scanning a keypad matrix. The KPP is capable of detecting, debouncing,
>> +and decoding one or multiple keys pressed simultaneously on a keypad.
>> +
>> +Required SoC Specific Properties:
>> +- compatible: Should be "fsl,imx-kpp".
>> +
>> +- reg: Physical base address of the KPP and length of memory mapped
>> + region.
>> +
>> +- interrupts: The KPP interrupt number to the CPU(s).
>> +
>> +- clocks: The clock provided by the SoC to the KPP. Some SoCs use ipg clock,
>> +others use dummy clock(The clock for the KPP is provided by the SoCs
>> +automatically).
>> +
>> +Required Board Specific Properties:
>> +- pinctrl-names: The definition can be found at
>> +pinctrl/pinctrl-bindings.txt.
>> +
>> +- pinctrl-0: The definition can be found at
>> +pinctrl/pinctrl-bindings.txt.
>> +
>> +- linux,keymap: The definition can be found at
>> +bindings/input/matrix-keymap.txt.
>> +
>> +Example:
>> +kpp: kpp at 73f94000 {
>> + compatible = "fsl,imx-kpp";
>> + reg = <0x73f94000 0x4000>;
>> + interrupts = <60>;
>> + clocks = <&clks 0>;
>> + pinctrl-names = "default";
>> + pinctrl-0 = <&pinctrl_kpp_1>;
>> + linux,keymap = <0x00000067 /* KEY_UP */
>> + 0x0001006c /* KEY_DOWN */
>> + 0x00020072 /* KEY_VOLUMEDOWN */
>> + 0x00030066 /* KEY_HOME */
>> + 0x0100006a /* KEY_RIGHT */
>> + 0x01010069 /* KEY_LEFT */
>> + 0x0102001c /* KEY_ENTER */
>> + 0x01030073 /* KEY_VOLUMEUP */
>> + 0x02000040 /* KEY_F6 */
>> + 0x02010042 /* KEY_F8 */
>> + 0x02020043 /* KEY_F9 */
>> + 0x02030044 /* KEY_F10 */
>> + 0x0300003b /* KEY_F1 */
>> + 0x0301003c /* KEY_F2 */
>> + 0x0302003d /* KEY_F3 */
>> + 0x03030074>; /* KEY_POWER */
>> +};
>> diff --git a/drivers/input/keyboard/imx_keypad.c b/drivers/input/keyboard/imx_keypad.c
>> index 6d150e3..e4f5c6a 100644
>> --- a/drivers/input/keyboard/imx_keypad.c
>> +++ b/drivers/input/keyboard/imx_keypad.c
>> @@ -20,6 +20,8 @@
>> #include <linux/jiffies.h>
>> #include <linux/kernel.h>
>> #include <linux/module.h>
>> +#include <linux/of.h>
>> +#include <linux/pinctrl/consumer.h>
>> #include <linux/platform_device.h>
>> #include <linux/slab.h>
>> #include <linux/timer.h>
>> @@ -414,17 +416,76 @@ open_err:
>> return -EIO;
>> }
>>
>> +static struct of_device_id imx_keypad_of_match[] = {
>> + { .compatible = "fsl,imx-kpp", },
>> + { /* sentinel */ }
>> +};
>> +MODULE_DEVICE_TABLE(of, imx_keypad_of_match);
>> +
>> +#ifdef CONFIG_OF
>> +static int imx_keypad_parse_dt(struct platform_device *pdev,
>> + uint32_t *keymap, int *keymap_size)
>> +{
>> + struct device_node *np = pdev->dev.of_node;
>> + struct device *dev = &pdev->dev;
>> + unsigned int proplen, i;
>> + const __be32 *prop;
>> +
>> + if (!np)
>> + return -ENODEV;
>> +
>> + prop = of_get_property(np, "linux,keymap", &proplen);
>> + if (!prop) {
>> + dev_err(dev, "OF: linux,keymap property not defined in %s\n",
>> + np->full_name);
>> + return -ENOENT;
>> + }
>> +
>> + if (proplen % sizeof(u32)) {
>> + dev_err(dev, "OF: Malformed keycode property in %s\n",
>> + np->full_name);
>> + return -EINVAL;
>> + }
>> +
>> + *keymap_size = proplen / sizeof(u32);
>> + if (*keymap_size > MAX_MATRIX_KEY_NUM) {
>> + dev_err(dev, "OF: linux,keymap size overflow\n");
>> + return -EINVAL;
>> + }
>> +
>> + for (i = 0; i < *keymap_size; i++)
>> + keymap[i] = be32_to_cpup(prop + i);
>
> All of this is implemented in matrix_keypad_build_keymap(), there is no
> need to do it again here.
The probe function of this driver needs the keymap information to set
keypad->rows_en_mask and keypad->cols_en_mask so that they can be used
later. I plan to change to parse the array keypad->keycodes[], which
is set by matrix_keypad_build_keymap(), to get the row/col mask. Do
you think this is acceptable?
>> +
>> + return 0;
>> +}
>> +#else
>> +static inline int imx_keypad_parse_dt(struct platform_device *pdev,
>> + uint32_t *keymap, int *keymap_size)
>> +{
>> + return -ENODEV;
>> +}
>> +#endif
>> +
>> static int imx_keypad_probe(struct platform_device *pdev)
>> {
>> const struct matrix_keymap_data *keymap_data = pdev->dev.platform_data;
>> struct imx_keypad *keypad;
>> struct input_dev *input_dev;
>> struct resource *res;
>> - int irq, error, i;
>> + int irq, error, i, keymap_size = 0;
>> + uint32_t keymap[MAX_MATRIX_KEY_NUM];
>> + struct pinctrl *pinctrl;
>>
>> - if (keymap_data == NULL) {
>> - dev_err(&pdev->dev, "no keymap defined\n");
>> - return -EINVAL;
>> + error = imx_keypad_parse_dt(pdev, keymap, &keymap_size);
>> + if (error) {
>> + if (keymap_data) {
>> + keymap_size = keymap_data->keymap_size;
>> + memcpy(keymap, keymap_data->keymap,
>> + keymap_size * sizeof(keymap[0]));
>
> I believe kernel-provided platform data should take precedence over
> DT-supplied data.
Ok. I will improve the logic here.
>
>> + } else {
>> + dev_err(&pdev->dev, "no keymap defined\n");
>> + return -EINVAL;
>> + }
>> }
>>
>> irq = platform_get_irq(pdev, 0);
>> @@ -473,6 +534,10 @@ static int imx_keypad_probe(struct platform_device *pdev)
>> goto failed_free_priv;
>> }
>>
>> + pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
>> + if (IS_ERR(pinctrl))
>> + dev_warn(&pdev->dev, "failed to get default pinctrl\n");
>> +
>
> No, please let core do this for you. Linus has a patch doing just that.
Ok, I will remove this.
>
>> keypad->clk = clk_get(&pdev->dev, NULL);
>> if (IS_ERR(keypad->clk)) {
>> dev_err(&pdev->dev, "failed to get keypad clock\n");
>> @@ -481,9 +546,9 @@ static int imx_keypad_probe(struct platform_device *pdev)
>> }
>>
>> /* Search for rows and cols enabled */
>> - for (i = 0; i < keymap_data->keymap_size; i++) {
>> - keypad->rows_en_mask |= 1 << KEY_ROW(keymap_data->keymap[i]);
>> - keypad->cols_en_mask |= 1 << KEY_COL(keymap_data->keymap[i]);
>> + for (i = 0; i < keymap_size; i++) {
>> + keypad->rows_en_mask |= 1 << KEY_ROW(keymap[i]);
>> + keypad->cols_en_mask |= 1 << KEY_COL(keymap[i]);
>> }
>>
>> if (keypad->rows_en_mask > ((1 << MAX_MATRIX_KEY_ROWS) - 1) ||
>> @@ -631,6 +696,7 @@ static struct platform_driver imx_keypad_driver = {
>> .name = "imx-keypad",
>> .owner = THIS_MODULE,
>> .pm = &imx_kbd_pm_ops,
>> + .of_match_table = of_match_ptr(imx_keypad_of_match),
>> },
>> .probe = imx_keypad_probe,
>> .remove = imx_keypad_remove,
>> --
>> 1.7.1
>>
>>
>
> Thanks.
>
> --
> Dmitry
--
Liu Ying
More information about the linux-arm-kernel
mailing list