leds: class: Don't expose color sysfs entry
PATCH 1/1 leds: class: Don't expose color sysfs entry
The commit c7d80059b086 ("leds: class: Store the color index in struct led_classdev") introduced a new sysfs entry "color" that is commonly created for the led classdev. Unfortunately, this conflicts with the "color" sysfs entry of already existing drivers such as Logitech HID or System76 ACPI drivers. The driver probe fails due to the conflict, hence it leads to a severe regression with the missing keyboard, for example.
This patch reverts partially the change in the commit above for removing the led class color sysfs entries again for addressing the regressions. The newly introduced led_classdev.color field is kept as it's already used by other driver.
---
This is a sort of v2 patch, as it turned out that the full revert leads to a build error.
Documentation/ABI/testing/sysfs-class-led | 9 --------- drivers/leds/led-class.c | 14 -------------- 2 files changed, 23 deletions(-)
Documentation/ABI/testing/sysfs-class-led +0 -9
@@ -59,15 +59,6 @@ Description: brightness. Reading this file when no hw brightness change event has happened will return an ENODATA error. -What: /sys/class/leds/<led>/color -Date: June 2023 -KernelVersion: 6.5 -Description: - Color of the LED. - - This is a read-only file. Reading this file returns the color - of the LED as a string (e.g: "red", "green", "multicolor"). - What: /sys/class/leds/<led>/trigger Date: March 2006 KernelVersion: 2.6.17
drivers/leds/led-class.c +0 -14
@@ -75,19 +75,6 @@ static ssize_t max_brightness_show(struct device *dev, }static DEVICE_ATTR_RO (max_brightness); -static ssize_t color_show (struct device *dev, -struct device_attribute *attr,char *buf) -{ -const char *color_text ="invalid" ; -struct led_classdev *led_cdev =dev_get_drvdata (dev); - -if (led_cdev->color <LED_COLOR_ID_MAX ) - color_text = led_colors[led_cdev->color]; - -return sysfs_emit (buf,"%s\n" , color_text); -} -static DEVICE_ATTR_RO (color); -#ifdef CONFIG_LEDS_TRIGGERS static BIN_ATTR(trigger, 0644, led_trigger_read, led_trigger_write, 0); static struct bin_attribute *led_trigger_bin_attrs[] = {
@@ -102,7 +89,6 @@ static const struct attribute_group led_trigger_group = {static struct attribute *led_class_attrs[] = { &dev_attr_brightness.attr, &dev_attr_max_brightness.attr, - &dev_attr_color.attr,NULL , };
Signature · 1 line
2.35.3
Hi,
Quoted 13 lines · Takashi Iwai
The commit
c7d80059b086("leds: class: Store the color index in structled_classdev") introduced a new sysfs entry "color" that is commonly created for the led classdev. Unfortunately, this conflicts with the "color" sysfs entry of already existing drivers such as Logitech HID or System76 ACPI drivers. The driver probe fails due to the conflict, hence it leads to a severe regression with the missing keyboard, for example.This patch reverts partially the change in the commit above for removing the led class color sysfs entries again for addressing the regressions. The newly introduced led_classdev.color field is kept as it's already used by other driver.
Quoted 7 lines
Fixes c7d80059b086 ("leds: class: Store the color index in struct led_classdev")Reported-by Johannes Penßel <[email protected]>Signed-off-by Takashi Iwai <[email protected]>
Thank you for taking care of this, patch looks good to me:
Regards,
Hans
Quoted 61 lines
---
This is a sort of v2 patch, as it turned out that the full revert leads to a build error.
Documentation/ABI/testing/sysfs-class-led | 9 --------- drivers/leds/led-class.c | 14 -------------- 2 files changed, 23 deletions(-)Documentation/ABI/testing/sysfs-class-led +0
-9@@ -59,15 +59,6 @@ Description: brightness. Reading this file when no hw brightness change event has happened will return an ENODATA error. -What: /sys/class/leds/<led>/color -Date: June 2023 -KernelVersion: 6.5 -Description: - Color of the LED. - - This is a read-only file. Reading this file returns the color - of the LED as a string (e.g: "red", "green", "multicolor"). - What: /sys/class/leds/<led>/trigger Date: March 2006 KernelVersion: 2.6.17drivers/leds/led-class.c +0
-14@@ -75,19 +75,6 @@ static ssize_t max_brightness_show(struct device *dev, }static DEVICE_ATTR_RO (max_brightness); -static ssize_t color_show (struct device *dev, -struct device_attribute *attr,char *buf) -{ -const char *color_text ="invalid" ; -struct led_classdev *led_cdev =dev_get_drvdata (dev); - -if (led_cdev->color <LED_COLOR_ID_MAX ) - color_text = led_colors[led_cdev->color]; - -return sysfs_emit (buf,"%s\n" , color_text); -} -static DEVICE_ATTR_RO (color); -#ifdef CONFIG_LEDS_TRIGGERS static BIN_ATTR(trigger, 0644, led_trigger_read, led_trigger_write, 0); static struct bin_attribute *led_trigger_bin_attrs[] = {@@ -102,7 +89,6 @@ static const struct attribute_group led_trigger_group = {static struct attribute *led_class_attrs[] = { &dev_attr_brightness.attr, &dev_attr_max_brightness.attr, - &dev_attr_color.attr,NULL , };
Quoted 9 lines · Takashi Iwai
The commit
c7d80059b086("leds: class: Store the color index in structled_classdev") introduced a new sysfs entry "color" that is commonly created for the led classdev. Unfortunately, this conflicts with the "color" sysfs entry of already existing drivers such as Logitech HID or System76 ACPI drivers. The driver probe fails due to the conflict, hence it leads to a severe regression with the missing keyboard, for example.[...]
Applied, thanks!
[1/1] leds: class: Don't expose color sysfs entry
commit: 8f2244c9af245ff72185c0473827125ee6b2d1a5
--
Lee Jones [李琼斯]