diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2019-09-17 18:40:42 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2019-09-17 18:40:42 -0700 |
commit | 4feaab05dc1eda3dbb57b097377766002e7a7cb9 (patch) | |
tree | fc7a877094032fd1e8d161c8782fecc73f70a8b6 /Documentation/leds/leds-class.rst | |
parent | Merge tag 'media/v5.4-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media (diff) | |
parent | leds: lm3532: Fix optional led-max-microamp prop error handling (diff) | |
download | linux-dev-4feaab05dc1eda3dbb57b097377766002e7a7cb9.tar.xz linux-dev-4feaab05dc1eda3dbb57b097377766002e7a7cb9.zip |
Merge tag 'leds-for-5.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds
Pull LED updates from Jacek Anaszewski:
"In this cycle we've finally managed to contribute the patch set
sorting out LED naming issues. Besides that there are many changes
scattered among various LED class drivers and triggers.
LED naming related improvements:
- add new 'function' and 'color' fwnode properties and deprecate
'label' property which has been frequently abused for conveying
vendor specific names that have been available in sysfs anyway
- introduce a set of standard LED_FUNCTION* definitions
- introduce a set of standard LED_COLOR_ID* definitions
- add a new {devm_}led_classdev_register_ext() API with the
capability of automatic LED name composition basing on the
properties available in the passed fwnode; the function is
backwards compatible in a sense that it uses 'label' data, if
present in the fwnode, for creating LED name
- add tools/leds/get_led_device_info.sh script for retrieving LED
vendor, product and bus names, if applicable; it also performs
basic validation of an LED name
- update following drivers and their DT bindings to use the new LED
registration API:
- leds-an30259a, leds-gpio, leds-as3645a, leds-aat1290, leds-cr0014114,
leds-lm3601x, leds-lm3692x, leds-lp8860, leds-lt3593, leds-sc27xx-blt
Other LED class improvements:
- replace {devm_}led_classdev_register() macros with inlines
- allow to call led_classdev_unregister() unconditionally
- switch to use fwnode instead of be stuck with OF one
LED triggers improvements:
- led-triggers:
- fix dereferencing of null pointer
- fix a memory leak bug
- ledtrig-gpio:
- GPIO 0 is valid
Drop superseeded apu2/3 support from leds-apu since for apu2+ a newer,
more complete driver exists, based on a generic driver for the AMD
SOCs gpio-controller, supporting LEDs as well other devices:
- drop profile field from priv data
- drop iosize field from priv data
- drop enum_apu_led_platform_types
- drop superseeded apu2/3 led support
- add pr_fmt prefix for better log output
- fix error message on probing failure
Other misc fixes and improvements to existing LED class drivers:
- leds-ns2, leds-max77650:
- add of_node_put() before return
- leds-pwm, leds-is31fl32xx:
- use struct_size() helper
- leds-lm3697, leds-lm36274, leds-lm3532:
- switch to use fwnode_property_count_uXX()
- leds-lm3532:
- fix brightness control for i2c mode
- change the define for the fs current register
- fixes for the driver for stability
- add full scale current configuration
- dt: Add property for full scale current.
- avoid potentially unpaired regulator calls
- move static keyword to the front of declarations
- fix optional led-max-microamp prop error handling
- leds-max77650:
- add of_node_put() before return
- add MODULE_ALIAS()
- Switch to fwnode property API
- leds-as3645a:
- fix misuse of strlcpy
- leds-netxbig:
- add of_node_put() in netxbig_leds_get_of_pdata()
- remove legacy board-file support
- leds-is31fl319x:
- simplify getting the adapter of a client
- leds-ti-lmu-common:
- fix coccinelle issue
- move static keyword to the front of declaration
- leds-syscon:
- use resource managed variant of device register
- leds-ktd2692:
- fix a typo in the name of a constant
- leds-lp5562:
- allow firmware files up to the maximum length
- leds-an30259a:
- fix typo
- leds-pca953x:
- include the right header"
* tag 'leds-for-5.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds: (72 commits)
leds: lm3532: Fix optional led-max-microamp prop error handling
led: triggers: Fix dereferencing of null pointer
leds: ti-lmu-common: Move static keyword to the front of declaration
leds: lm3532: Move static keyword to the front of declarations
leds: trigger: gpio: GPIO 0 is valid
leds: pwm: Use struct_size() helper
leds: is31fl32xx: Use struct_size() helper
leds: ti-lmu-common: Fix coccinelle issue in TI LMU
leds: lm3532: Avoid potentially unpaired regulator calls
leds: syscon: Use resource managed variant of device register
leds: Replace {devm_}led_classdev_register() macros with inlines
leds: Allow to call led_classdev_unregister() unconditionally
leds: lm3532: Add full scale current configuration
dt: lm3532: Add property for full scale current.
leds: lm3532: Fixes for the driver for stability
leds: lm3532: Change the define for the fs current register
leds: lm3532: Fix brightness control for i2c mode
leds: Switch to use fwnode instead of be stuck with OF one
leds: max77650: Switch to fwnode property API
led: triggers: Fix a memory leak bug
...
Diffstat (limited to 'Documentation/leds/leds-class.rst')
-rw-r--r-- | Documentation/leds/leds-class.rst | 70 |
1 files changed, 67 insertions, 3 deletions
diff --git a/Documentation/leds/leds-class.rst b/Documentation/leds/leds-class.rst index df0120a1ee3c..a0708d3f3d0b 100644 --- a/Documentation/leds/leds-class.rst +++ b/Documentation/leds/leds-class.rst @@ -43,9 +43,73 @@ LED Device Naming Is currently of the form: - "devicename:colour:function" - -There have been calls for LED properties such as colour to be exported as + "devicename:color:function" + +- devicename: + it should refer to a unique identifier created by the kernel, + like e.g. phyN for network devices or inputN for input devices, rather + than to the hardware; the information related to the product and the bus + to which given device is hooked is available in sysfs and can be + retrieved using get_led_device_info.sh script from tools/leds; generally + this section is expected mostly for LEDs that are somehow associated with + other devices. + +- color: + one of LED_COLOR_ID_* definitions from the header + include/dt-bindings/leds/common.h. + +- function: + one of LED_FUNCTION_* definitions from the header + include/dt-bindings/leds/common.h. + +If required color or function is missing, please submit a patch +to linux-leds@vger.kernel.org. + +It is possible that more than one LED with the same color and function will +be required for given platform, differing only with an ordinal number. +In this case it is preferable to just concatenate the predefined LED_FUNCTION_* +name with required "-N" suffix in the driver. fwnode based drivers can use +function-enumerator property for that and then the concatenation will be handled +automatically by the LED core upon LED class device registration. + +LED subsystem has also a protection against name clash, that may occur +when LED class device is created by a driver of hot-pluggable device and +it doesn't provide unique devicename section. In this case numerical +suffix (e.g. "_1", "_2", "_3" etc.) is added to the requested LED class +device name. + +There might be still LED class drivers around using vendor or product name +for devicename, but this approach is now deprecated as it doesn't convey +any added value. Product information can be found in other places in sysfs +(see tools/leds/get_led_device_info.sh). + +Examples of proper LED names: + + - "red:disk" + - "white:flash" + - "red:indicator" + - "phy1:green:wlan" + - "phy3::wlan" + - ":kbd_backlight" + - "input5::kbd_backlight" + - "input3::numlock" + - "input3::scrolllock" + - "input3::capslock" + - "mmc1::status" + - "white:status" + +get_led_device_info.sh script can be used for verifying if the LED name +meets the requirements pointed out here. It performs validation of the LED class +devicename sections and gives hints on expected value for a section in case +the validation fails for it. So far the script supports validation +of associations between LEDs and following types of devices: + + - input devices + - ieee80211 compliant USB devices + +The script is open to extensions. + +There have been calls for LED properties such as color to be exported as individual led class attributes. As a solution which doesn't incur as much overhead, I suggest these become part of the device name. The naming scheme above leaves scope for further attributes should they be needed. If sections |