Mauro Carvalho Chehab | 8dab919 | 2019-06-28 09:20:20 -0300 | [diff] [blame] | 1 | ======================== |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 2 | LED handling under Linux |
| 3 | ======================== |
| 4 | |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 5 | In its simplest form, the LED class just allows control of LEDs from |
Corentin Chary | 5f634c6 | 2009-08-28 12:56:45 +0000 | [diff] [blame] | 6 | userspace. LEDs appear in /sys/class/leds/. The maximum brightness of the |
| 7 | LED is defined in max_brightness file. The brightness file will set the brightness |
| 8 | of the LED (taking a value 0-max_brightness). Most LEDs don't have hardware |
| 9 | brightness support so will just be turned on for non-zero brightness settings. |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 10 | |
| 11 | The class also introduces the optional concept of an LED trigger. A trigger |
| 12 | is a kernel based source of led events. Triggers can either be simple or |
| 13 | complex. A simple trigger isn't configurable and is designed to slot into |
Stephan Linz | 86ab168 | 2016-06-24 19:16:53 +0200 | [diff] [blame] | 14 | existing subsystems with minimal additional code. Examples are the disk-activity, |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 15 | nand-disk and sharpsl-charge triggers. With led triggers disabled, the code |
| 16 | optimises away. |
| 17 | |
Will Deacon | 806654a | 2018-11-19 11:02:45 +0000 | [diff] [blame] | 18 | Complex triggers while available to all LEDs have LED specific |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 19 | parameters and work on a per LED basis. The timer trigger is an example. |
Németh Márton | 0013b23 | 2008-03-09 20:54:37 +0000 | [diff] [blame] | 20 | The timer trigger will periodically change the LED brightness between |
| 21 | LED_OFF and the current brightness setting. The "on" and "off" time can |
| 22 | be specified via /sys/class/leds/<device>/delay_{on,off} in milliseconds. |
| 23 | You can change the brightness value of a LED independently of the timer |
| 24 | trigger. However, if you set the brightness value to LED_OFF it will |
| 25 | also disable the timer trigger. |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 26 | |
| 27 | You can change triggers in a similar manner to the way an IO scheduler |
| 28 | is chosen (via /sys/class/leds/<device>/trigger). Trigger specific |
| 29 | parameters can appear in /sys/class/leds/<device> once a given trigger is |
| 30 | selected. |
| 31 | |
| 32 | |
| 33 | Design Philosophy |
| 34 | ================= |
| 35 | |
| 36 | The underlying design philosophy is simplicity. LEDs are simple devices |
| 37 | and the aim is to keep a small amount of code giving as much functionality |
| 38 | as possible. Please keep this in mind when suggesting enhancements. |
| 39 | |
| 40 | |
| 41 | LED Device Naming |
| 42 | ================= |
| 43 | |
| 44 | Is currently of the form: |
| 45 | |
Jacek Anaszewski | bb4e9af | 2019-06-09 20:19:04 +0200 | [diff] [blame] | 46 | "devicename:color:function" |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 47 | |
Jacek Anaszewski | bb4e9af | 2019-06-09 20:19:04 +0200 | [diff] [blame] | 48 | - devicename: |
| 49 | it should refer to a unique identifier created by the kernel, |
| 50 | like e.g. phyN for network devices or inputN for input devices, rather |
| 51 | than to the hardware; the information related to the product and the bus |
| 52 | to which given device is hooked is available in sysfs and can be |
| 53 | retrieved using get_led_device_info.sh script from tools/leds; generally |
| 54 | this section is expected mostly for LEDs that are somehow associated with |
| 55 | other devices. |
| 56 | |
| 57 | - color: |
| 58 | one of LED_COLOR_ID_* definitions from the header |
| 59 | include/dt-bindings/leds/common.h. |
| 60 | |
| 61 | - function: |
| 62 | one of LED_FUNCTION_* definitions from the header |
| 63 | include/dt-bindings/leds/common.h. |
| 64 | |
| 65 | If required color or function is missing, please submit a patch |
| 66 | to linux-leds@vger.kernel.org. |
| 67 | |
| 68 | It is possible that more than one LED with the same color and function will |
| 69 | be required for given platform, differing only with an ordinal number. |
| 70 | In this case it is preferable to just concatenate the predefined LED_FUNCTION_* |
| 71 | name with required "-N" suffix in the driver. fwnode based drivers can use |
| 72 | function-enumerator property for that and then the concatenation will be handled |
| 73 | automatically by the LED core upon LED class device registration. |
| 74 | |
| 75 | LED subsystem has also a protection against name clash, that may occur |
| 76 | when LED class device is created by a driver of hot-pluggable device and |
| 77 | it doesn't provide unique devicename section. In this case numerical |
| 78 | suffix (e.g. "_1", "_2", "_3" etc.) is added to the requested LED class |
| 79 | device name. |
| 80 | |
| 81 | There might be still LED class drivers around using vendor or product name |
| 82 | for devicename, but this approach is now deprecated as it doesn't convey |
| 83 | any added value. Product information can be found in other places in sysfs |
| 84 | (see tools/leds/get_led_device_info.sh). |
| 85 | |
| 86 | Examples of proper LED names: |
| 87 | |
| 88 | - "red:disk" |
| 89 | - "white:flash" |
| 90 | - "red:indicator" |
| 91 | - "phy1:green:wlan" |
| 92 | - "phy3::wlan" |
| 93 | - ":kbd_backlight" |
| 94 | - "input5::kbd_backlight" |
| 95 | - "input3::numlock" |
| 96 | - "input3::scrolllock" |
| 97 | - "input3::capslock" |
| 98 | - "mmc1::status" |
| 99 | - "white:status" |
| 100 | |
| 101 | get_led_device_info.sh script can be used for verifying if the LED name |
| 102 | meets the requirements pointed out here. It performs validation of the LED class |
| 103 | devicename sections and gives hints on expected value for a section in case |
| 104 | the validation fails for it. So far the script supports validation |
| 105 | of associations between LEDs and following types of devices: |
| 106 | |
| 107 | - input devices |
| 108 | - ieee80211 compliant USB devices |
| 109 | |
| 110 | The script is open to extensions. |
| 111 | |
| 112 | There have been calls for LED properties such as color to be exported as |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 113 | individual led class attributes. As a solution which doesn't incur as much |
| 114 | overhead, I suggest these become part of the device name. The naming scheme |
Richard Purdie | 6c152be | 2007-10-31 15:00:07 +0100 | [diff] [blame] | 115 | above leaves scope for further attributes should they be needed. If sections |
| 116 | of the name don't apply, just leave that section blank. |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 117 | |
| 118 | |
Jacek Anaszewski | 648da8f | 2015-10-07 11:10:44 +0200 | [diff] [blame] | 119 | Brightness setting API |
| 120 | ====================== |
| 121 | |
| 122 | LED subsystem core exposes following API for setting brightness: |
| 123 | |
Mauro Carvalho Chehab | 8dab919 | 2019-06-28 09:20:20 -0300 | [diff] [blame] | 124 | - led_set_brightness: |
| 125 | it is guaranteed not to sleep, passing LED_OFF stops |
Jacek Anaszewski | 648da8f | 2015-10-07 11:10:44 +0200 | [diff] [blame] | 126 | blinking, |
Mauro Carvalho Chehab | 8dab919 | 2019-06-28 09:20:20 -0300 | [diff] [blame] | 127 | |
| 128 | - led_set_brightness_sync: |
| 129 | for use cases when immediate effect is desired - |
Jacek Anaszewski | 648da8f | 2015-10-07 11:10:44 +0200 | [diff] [blame] | 130 | it can block the caller for the time required for accessing |
| 131 | device registers and can sleep, passing LED_OFF stops hardware |
| 132 | blinking, returns -EBUSY if software blink fallback is enabled. |
| 133 | |
| 134 | |
Hans de Goede | 0cb8eb3 | 2017-01-29 14:42:52 +0100 | [diff] [blame] | 135 | LED registration API |
| 136 | ==================== |
| 137 | |
| 138 | A driver wanting to register a LED classdev for use by other drivers / |
| 139 | userspace needs to allocate and fill a led_classdev struct and then call |
Mauro Carvalho Chehab | 8dab919 | 2019-06-28 09:20:20 -0300 | [diff] [blame] | 140 | `[devm_]led_classdev_register`. If the non devm version is used the driver |
Hans de Goede | 0cb8eb3 | 2017-01-29 14:42:52 +0100 | [diff] [blame] | 141 | must call led_classdev_unregister from its remove function before |
| 142 | free-ing the led_classdev struct. |
| 143 | |
| 144 | If the driver can detect hardware initiated brightness changes and thus |
| 145 | wants to have a brightness_hw_changed attribute then the LED_BRIGHT_HW_CHANGED |
| 146 | flag must be set in flags before registering. Calling |
| 147 | led_classdev_notify_brightness_hw_changed on a classdev not registered with |
| 148 | the LED_BRIGHT_HW_CHANGED flag is a bug and will trigger a WARN_ON. |
| 149 | |
Márton Németh | 4c79141 | 2007-10-31 15:07:12 +0100 | [diff] [blame] | 150 | Hardware accelerated blink of LEDs |
| 151 | ================================== |
| 152 | |
| 153 | Some LEDs can be programmed to blink without any CPU interaction. To |
| 154 | support this feature, a LED driver can optionally implement the |
Johannes Berg | 5ada28b | 2010-11-11 14:05:21 -0800 | [diff] [blame] | 155 | blink_set() function (see <linux/leds.h>). To set an LED to blinking, |
Bryan Wu | ee31892 | 2011-11-04 11:22:29 -0700 | [diff] [blame] | 156 | however, it is better to use the API function led_blink_set(), as it |
| 157 | will check and implement software fallback if necessary. |
Márton Németh | 4c79141 | 2007-10-31 15:07:12 +0100 | [diff] [blame] | 158 | |
Tony Makkiel | 7cfe749 | 2016-05-18 17:22:45 +0100 | [diff] [blame] | 159 | To turn off blinking, use the API function led_brightness_set() |
| 160 | with brightness value LED_OFF, which should stop any software |
Johannes Berg | 5ada28b | 2010-11-11 14:05:21 -0800 | [diff] [blame] | 161 | timers that may have been required for blinking. |
| 162 | |
| 163 | The blink_set() function should choose a user friendly blinking value |
Mauro Carvalho Chehab | 8dab919 | 2019-06-28 09:20:20 -0300 | [diff] [blame] | 164 | if it is called with `*delay_on==0` && `*delay_off==0` parameters. In this |
Johannes Berg | 5ada28b | 2010-11-11 14:05:21 -0800 | [diff] [blame] | 165 | case the driver should give back the chosen value through delay_on and |
| 166 | delay_off parameters to the leds subsystem. |
Márton Németh | 4c79141 | 2007-10-31 15:07:12 +0100 | [diff] [blame] | 167 | |
Németh Márton | 0013b23 | 2008-03-09 20:54:37 +0000 | [diff] [blame] | 168 | Setting the brightness to zero with brightness_set() callback function |
| 169 | should completely turn off the LED and cancel the previously programmed |
| 170 | hardware blinking function, if any. |
Márton Németh | 4c79141 | 2007-10-31 15:07:12 +0100 | [diff] [blame] | 171 | |
| 172 | |
Richard Purdie | 75c1d31 | 2006-03-31 02:31:03 -0800 | [diff] [blame] | 173 | Known Issues |
| 174 | ============ |
| 175 | |
| 176 | The LED Trigger core cannot be a module as the simple trigger functions |
| 177 | would cause nightmare dependency issues. I see this as a minor issue |
| 178 | compared to the benefits the simple trigger functionality brings. The |
| 179 | rest of the LED subsystem can be modular. |