Skip to contentSkip to conversation

leds-gpio: just ignore invalid GPIOs

PATCH In review David Brownell 2 participants 3 messages Mar 6, 2009 LED

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 leds-gpio: just ignore invalid GPIOs

From: David Brownell <[email protected]>

Sometimes it's awkward to make sure that the array in the platform_data handed to the leds-gpio driver has only valid data ... some leds may not be always available, and coping with that currently requires patching or rebuilding the array.

This patch fixes that by making it be OK to pass an invalid GPIO (such as "-EINVAL") ... such table entries are skipped.

---
Noted on OMAP3 Beagle board ... one of the three programmable LEDs is driven by a twl4030 GPIO, which might not be configured. (In fact it's a PWM-capable LED, but the PWM support isn't yet available.)

 drivers/leds/leds-gpio.c |   12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/leds/leds-gpio.c +11 -1
@@ -90,13 +90,19 @@ static int gpio_led_probe(struct platfor
 		cur_led = &pdata->leds[i];
 		led_dat = &leds_data[i];
 
+		/* skip leds that aren't available */
+		led_dat->gpio = cur_led->gpio;
+		if (!gpio_is_valid(led_dat->gpio)) {
+			dev_dbg(&pdev->dev, "skipping %s\n", cur_led->name);
+			continue;
+		}
+
 		ret = gpio_request(cur_led->gpio, cur_led->name);
 		if (ret < 0)
 			goto err;
 
 		led_dat->cdev.name = cur_led->name;
 		led_dat->cdev.default_trigger = cur_led->default_trigger;
-		led_dat->gpio = cur_led->gpio;
 		led_dat->can_sleep = gpio_cansleep(cur_led->gpio);
 		led_dat->active_low = cur_led->active_low;
 		if (pdata->gpio_blink_set) {
@@ -125,6 +131,8 @@ static int gpio_led_probe(struct platfor
 err:
 	if (i > 0) {
 		for (i = i - 1; i >= 0; i--) {
+			if (!gpio_is_valid(leds_data[i].gpio))
+				continue;
 			led_classdev_unregister(&leds_data[i].cdev);
 			cancel_work_sync(&leds_data[i].work);
 			gpio_free(leds_data[i].gpio);
@@ -145,6 +153,8 @@ static int __devexit gpio_led_remove(str
 	leds_data = platform_get_drvdata(pdev);
 
 	for (i = 0; i < pdata->num_leds; i++) {
+		if (!gpio_is_valid(leds_data[i].gpio))
+			continue;
 		led_classdev_unregister(&leds_data[i].cdev);
 		cancel_work_sync(&leds_data[i].work);
 		gpio_free(leds_data[i].gpio);
#1 lore raw
Quoted 54 lines · On Thu, 5 Mar 2009 16:46:44 -0800 David Brownell <[email protected]> wrote:

From: David Brownell <[email protected]>

Sometimes it's awkward to make sure that the array in the platform_data handed to the leds-gpio driver has only valid data ... some leds may not be always available, and coping with that currently requires patching or rebuilding the array.

This patch fixes that by making it be OK to pass an invalid GPIO (such as "-EINVAL") ... such table entries are skipped.

---
Noted on OMAP3 Beagle board ... one of the three programmable LEDs is driven by a twl4030 GPIO, which might not be configured. (In fact it's a PWM-capable LED, but the PWM support isn't yet available.)

 drivers/leds/leds-gpio.c |   12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/leds/leds-gpio.c +9 -1
@@ -90,13 +90,19 @@ static int gpio_led_probe(struct platfor
 		cur_led = &pdata->leds[i];
 		led_dat = &leds_data[i];
 
+		/* skip leds that aren't available */
+		led_dat->gpio = cur_led->gpio;
+		if (!gpio_is_valid(led_dat->gpio)) {
+			dev_dbg(&pdev->dev, "skipping %s\n", cur_led->name);
+			continue;
+		}
+
 		ret = gpio_request(cur_led->gpio, cur_led->name);
 		if (ret < 0)
 			goto err;
 
 		led_dat->cdev.name = cur_led->name;
 		led_dat->cdev.default_trigger = cur_led->default_trigger;
-		led_dat->gpio = cur_led->gpio;
 		led_dat->can_sleep = gpio_cansleep(cur_led->gpio);
 		led_dat->active_low = cur_led->active_low;
 		if (pdata->gpio_blink_set) {
@@ -125,6 +131,8 @@ static int gpio_led_probe(struct platfor
 err:
 	if (i > 0) {
 		for (i = i - 1; i >= 0; i--) {
+			if (!gpio_is_valid(leds_data[i].gpio))
+				continue;
 			led_classdev_unregister(&leds_data[i].cdev);
 			cancel_work_sync(&leds_data[i].work);
 			gpio_free(leds_data[i].gpio);

drivers/leds/leds-gpio.c static int __devexit gpio_led_remove(str · line 160

Quoted 8 lines
drivers/leds/leds-gpio.c +2 -0
@@ -145,6 +153,8 @@ static int __devexit gpio_led_remove(str
 	leds_data = platform_get_drvdata(pdev);
 
 	for (i = 0; i < pdata->num_leds; i++) {
+		if (!gpio_is_valid(leds_data[i].gpio))
+			continue;
 		led_classdev_unregister(&leds_data[i].cdev);
 		cancel_work_sync(&leds_data[i].work);
 		gpio_free(leds_data[i].gpio);

All this code has changed a lot in linux.next. Nothing applies and it goes beyond my fix-it-up preparedness.

#2 lore raw
Quoted 2 lines · Andrew Morton

All this code has changed a lot in linux.next.  Nothing applies and it goes beyond my fix-it-up preparedness.

Well that's just no fun. :(

Unless someone else wants to handle that, I'm just going to leave this version in my patch queue until those changes come down from mainline and break it. An updated version would come sometime later.

- Dave