2013-07-07 10:10:38

by Samuel Thibault

[permalink] [raw]
Subject: [PATCH] Route kbd LEDs through the generic LEDs layer

This permits to reassign keyboard LEDs to something else than keyboard "leds"
state, by adding keyboard led and modifier triggers connected to a series
of VT input LEDs, themselves connected to VT input triggers, which
per-input device LEDs use by default. Userland can thus easily change the LED
behavior of (a priori) all input devices, or of particular input devices.

This also permits to fix #7063 from userland by using a modifier to implement
proper CapsLock behavior and have the keyboard caps lock led show that modifier
state.

Signed-off-by: Samuel Thibault <[email protected]>
[[email protected]: Rebased to 3.2-rc1 or so, cleaned up some includes, and fixed some constants]
Signed-off-by: Evan Broder <[email protected]>

diff --exclude .svn -ur linux-3.10-orig/Documentation/leds/leds-class.txt linux-3.10-perso/Documentation/leds/leds-class.txt
--- linux-3.10-orig/Documentation/leds/leds-class.txt 2013-04-29 13:27:16.959258613 +0200
+++ linux-3.10-perso/Documentation/leds/leds-class.txt 2013-07-01 23:51:39.229318781 +0200
@@ -2,9 +2,6 @@
LED handling under Linux
========================

-If you're reading this and thinking about keyboard leds, these are
-handled by the input subsystem and the led class is *not* needed.
-
In its simplest form, the LED class just allows control of LEDs from
userspace. LEDs appear in /sys/class/leds/. The maximum brightness of the
LED is defined in max_brightness file. The brightness file will set the brightness
diff --exclude .svn -ur linux-3.10-orig/drivers/input/input.c linux-3.10-perso/drivers/input/input.c
--- linux-3.10-orig/drivers/input/input.c 2013-04-29 13:27:16.959258613 +0200
+++ linux-3.10-perso/drivers/input/input.c 2013-07-01 23:51:39.233318421 +0200
@@ -28,6 +28,7 @@
#include <linux/mutex.h>
#include <linux/rcupdate.h>
#include "input-compat.h"
+#include "leds.h"

MODULE_AUTHOR("Vojtech Pavlik <[email protected]>");
MODULE_DESCRIPTION("Input core");
@@ -708,6 +709,11 @@
handle->open = 0;

spin_unlock_irq(&dev->event_lock);
+
+#ifdef CONFIG_INPUT_LEDS
+ if (is_event_supported(EV_LED, dev->evbit, EV_MAX))
+ input_led_disconnect(dev);
+#endif
}

/**
@@ -2092,6 +2098,11 @@

list_add_tail(&dev->node, &input_dev_list);

+#ifdef CONFIG_INPUT_LEDS
+ if (is_event_supported(EV_LED, dev->evbit, EV_MAX))
+ input_led_connect(dev);
+#endif
+
list_for_each_entry(handler, &input_handler_list, node)
input_attach_handler(dev, handler);

diff --exclude .svn -ur linux-3.10-orig/drivers/input/Kconfig linux-3.10-perso/drivers/input/Kconfig
--- linux-3.10-orig/drivers/input/Kconfig 2013-04-29 13:32:29.997497163 +0200
+++ linux-3.10-perso/drivers/input/Kconfig 2013-07-01 23:51:39.233318421 +0200
@@ -178,6 +178,15 @@

source "drivers/input/keyboard/Kconfig"

+config INPUT_LEDS
+ tristate "LED Support"
+ depends on LEDS_CLASS
+ select LEDS_TRIGGERS
+ default y
+ help
+ This option enables support for the LEDs on keyboard managed
+ by the input layer.
+
source "drivers/input/mouse/Kconfig"

source "drivers/input/joystick/Kconfig"
diff --exclude .svn -ur linux-3.10-orig/drivers/input/Makefile linux-3.10-perso/drivers/input/Makefile
--- linux-3.10-orig/drivers/input/Makefile 2013-04-29 13:27:16.959258613 +0200
+++ linux-3.10-perso/drivers/input/Makefile 2013-07-01 23:51:39.233318421 +0200
@@ -18,6 +18,7 @@
obj-$(CONFIG_INPUT_EVBUG) += evbug.o

obj-$(CONFIG_INPUT_KEYBOARD) += keyboard/
+obj-$(CONFIG_INPUT_LEDS) += leds.o
obj-$(CONFIG_INPUT_MOUSE) += mouse/
obj-$(CONFIG_INPUT_JOYSTICK) += joystick/
obj-$(CONFIG_INPUT_TABLET) += tablet/
diff --exclude .svn -ur linux-3.10-orig/drivers/leds/Kconfig linux-3.10-perso/drivers/leds/Kconfig
--- linux-3.10-orig/drivers/leds/Kconfig 2013-07-01 01:56:00.335874214 +0200
+++ linux-3.10-perso/drivers/leds/Kconfig 2013-07-01 23:51:39.233318421 +0200
@@ -11,9 +11,6 @@
Say Y to enable Linux LED support. This allows control of supported
LEDs from both userspace and optionally, by kernel events (triggers).

- This is not related to standard keyboard LEDs which are controlled
- via the input system.
-
if NEW_LEDS

config LEDS_CLASS
diff --exclude .svn -ur linux-3.10-orig/drivers/tty/Kconfig linux-3.10-perso/drivers/tty/Kconfig
--- linux-3.10-orig/drivers/tty/Kconfig 2013-04-29 13:32:36.353298705 +0200
+++ linux-3.10-perso/drivers/tty/Kconfig 2013-07-01 23:51:39.237318056 +0200
@@ -13,6 +13,10 @@
bool "Virtual terminal" if EXPERT
depends on !S390 && !UML
select INPUT
+ select NEW_LEDS
+ select LEDS_CLASS
+ select LEDS_TRIGGERS
+ select INPUT_LEDS
default y
---help---
If you say Y here, you will get support for terminal devices with
diff --exclude .svn -ur linux-3.10-orig/drivers/tty/vt/keyboard.c linux-3.10-perso/drivers/tty/vt/keyboard.c
--- linux-3.10-orig/drivers/tty/vt/keyboard.c 2013-04-29 13:32:36.681288463 +0200
+++ linux-3.10-perso/drivers/tty/vt/keyboard.c 2013-07-01 23:51:39.237318056 +0200
@@ -33,6 +33,7 @@
#include <linux/string.h>
#include <linux/init.h>
#include <linux/slab.h>
+#include <linux/leds.h>

#include <linux/kbd_kern.h>
#include <linux/kbd_diacr.h>
@@ -130,6 +131,7 @@
static int shift_state = 0;

static unsigned char ledstate = 0xff; /* undefined */
+static unsigned char lockstate = 0xff; /* undefined */
static unsigned char ledioctl;

static struct ledptr {
@@ -967,6 +969,32 @@
}
}

+/* We route VT keyboard "leds" through triggers */
+static void kbd_ledstate_trigger_activate(struct led_classdev *cdev);
+static struct led_trigger ledtrig_ledstate[] = {
+#define DEFINE_LEDSTATE_TRIGGER(kbd_led, nam) \
+ [kbd_led] = { .name = nam, .activate = kbd_ledstate_trigger_activate, }
+ DEFINE_LEDSTATE_TRIGGER(VC_SCROLLOCK, "scrollock"),
+ DEFINE_LEDSTATE_TRIGGER(VC_NUMLOCK, "numlock"),
+ DEFINE_LEDSTATE_TRIGGER(VC_CAPSLOCK, "capslock"),
+ DEFINE_LEDSTATE_TRIGGER(VC_KANALOCK, "kanalock"),
+#undef DEFINE_LEDSTATE_TRIGGER
+};
+static void kbd_lockstate_trigger_activate(struct led_classdev *cdev);
+static struct led_trigger ledtrig_lockstate[] = {
+#define DEFINE_LOCKSTATE_TRIGGER(kbd_led, nam) \
+ [kbd_led] = { .name = nam, .activate = kbd_lockstate_trigger_activate, }
+ DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLOCK, "shiftlock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_ALTGRLOCK, "altgrlock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_CTRLLOCK, "ctrllock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_ALTLOCK, "altlock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLLOCK, "shiftllock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTRLOCK, "shiftrlock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_CTRLLLOCK, "ctrlllock"),
+ DEFINE_LOCKSTATE_TRIGGER(VC_CTRLRLOCK, "ctrlrlock"),
+#undef DEFINE_LOCKSTATE_TRIGGER
+};
+
/*
* The leds display either (i) the status of NumLock, CapsLock, ScrollLock,
* or (ii) whatever pattern of lights people want to show using KDSETLED,
@@ -1002,6 +1030,7 @@

leds = kbd->ledflagstate;

+ /* TODO: should be replaced by a LED trigger */
if (kbd->ledmode == LED_SHOW_MEM) {
for (i = 0; i < 3; i++)
if (ledptrs[i].valid) {
@@ -1014,18 +1043,24 @@
return leds;
}

-static int kbd_update_leds_helper(struct input_handle *handle, void *data)
+/* Called on trigger connection, to set initial state */
+static void kbd_ledstate_trigger_activate(struct led_classdev *cdev)
{
- unsigned char leds = *(unsigned char *)data;
+ struct led_trigger *trigger = cdev->trigger;
+ int led = trigger - ledtrig_ledstate;

- if (test_bit(EV_LED, handle->dev->evbit)) {
- input_inject_event(handle, EV_LED, LED_SCROLLL, !!(leds & 0x01));
- input_inject_event(handle, EV_LED, LED_NUML, !!(leds & 0x02));
- input_inject_event(handle, EV_LED, LED_CAPSL, !!(leds & 0x04));
- input_inject_event(handle, EV_SYN, SYN_REPORT, 0);
- }
+ tasklet_disable(&keyboard_tasklet);
+ led_trigger_event(trigger, ledstate & (1 << led) ? LED_FULL : LED_OFF);
+ tasklet_enable(&keyboard_tasklet);
+}
+static void kbd_lockstate_trigger_activate(struct led_classdev *cdev)
+{
+ struct led_trigger *trigger = cdev->trigger;
+ int led = trigger - ledtrig_lockstate;

- return 0;
+ tasklet_disable(&keyboard_tasklet);
+ led_trigger_event(trigger, lockstate & (1 << led) ? LED_FULL : LED_OFF);
+ tasklet_enable(&keyboard_tasklet);
}

/**
@@ -1120,10 +1155,24 @@
spin_unlock_irqrestore(&led_lock, flags);

if (leds != ledstate) {
- input_handler_for_each_handle(&kbd_handler, &leds,
- kbd_update_leds_helper);
+ int i;
+ for (i = 0; i < ARRAY_SIZE(ledtrig_ledstate); i++)
+ if ((leds ^ ledstate) & (1 << i))
+ led_trigger_event(&ledtrig_ledstate[i],
+ leds & (1 << i)
+ ? LED_FULL : LED_OFF);
ledstate = leds;
}
+
+ if (kbd->lockstate != lockstate) {
+ int i;
+ for (i = 0; i < ARRAY_SIZE(ledtrig_lockstate); i++)
+ if ((kbd->lockstate ^ lockstate) & (1 << i))
+ led_trigger_event(&ledtrig_lockstate[i],
+ kbd->lockstate & (1 << i)
+ ? LED_FULL : LED_OFF);
+ lockstate = kbd->lockstate;
+ }
}

DECLARE_TASKLET_DISABLED(keyboard_tasklet, kbd_bh, 0);
@@ -1461,20 +1510,6 @@
kfree(handle);
}

-/*
- * Start keyboard handler on the new keyboard by refreshing LED state to
- * match the rest of the system.
- */
-static void kbd_start(struct input_handle *handle)
-{
- tasklet_disable(&keyboard_tasklet);
-
- if (ledstate != 0xff)
- kbd_update_leds_helper(handle, &ledstate);
-
- tasklet_enable(&keyboard_tasklet);
-}
-
static const struct input_device_id kbd_ids[] = {
{
.flags = INPUT_DEVICE_ID_MATCH_EVBIT,
@@ -1496,7 +1531,6 @@
.match = kbd_match,
.connect = kbd_connect,
.disconnect = kbd_disconnect,
- .start = kbd_start,
.name = "kbd",
.id_table = kbd_ids,
};
@@ -1520,6 +1554,11 @@
if (error)
return error;

+ for (i = 0; i < ARRAY_SIZE(ledtrig_ledstate); i++)
+ led_trigger_register(&ledtrig_ledstate[i]);
+ for (i = 0; i < ARRAY_SIZE(ledtrig_lockstate); i++)
+ led_trigger_register(&ledtrig_lockstate[i]);
+
tasklet_enable(&keyboard_tasklet);
tasklet_schedule(&keyboard_tasklet);

diff --exclude .svn -ur linux-3.10-orig/include/linux/input.h linux-3.10-perso/include/linux/input.h
--- linux-3.10-orig/include/linux/input.h 2013-04-29 13:27:16.959258613 +0200
+++ linux-3.10-perso/include/linux/input.h 2013-07-01 23:51:39.241317685 +0200
@@ -164,6 +164,8 @@
unsigned long snd[BITS_TO_LONGS(SND_CNT)];
unsigned long sw[BITS_TO_LONGS(SW_CNT)];

+ struct led_classdev *leds;
+
int (*open)(struct input_dev *dev);
void (*close)(struct input_dev *dev);
int (*flush)(struct input_dev *dev, struct file *file);
--- linux-3.10-orig/drivers/input/leds.h 1970-01-01 01:00:00.000000000 +0100
+++ linux-3.10-perso/drivers/input/leds.h 2011-11-14 02:24:26.738456456 +0100
@@ -0,0 +1,14 @@
+/*
+ * LED support for the input layer
+ *
+ * Copyright 2010-2013 Samuel Thibault <[email protected]>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/input.h>
+
+extern int input_led_connect(struct input_dev *dev);
+extern void input_led_disconnect(struct input_dev *dev);
--- linux-3.10-orig/drivers/input/leds.c 1970-01-01 01:00:00.000000000 +0100
+++ linux-3.10-perso/drivers/input/leds.c 2011-11-14 03:23:07.578469395 +0100
@@ -0,0 +1,243 @@
+/*
+ * LED support for the input layer
+ *
+ * Copyright 2010-2013 Samuel Thibault <[email protected]>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/leds.h>
+#include <linux/input.h>
+
+/*
+ * Keyboard LEDs are propagated by default like the following example:
+ *
+ * VT keyboard numlock trigger
+ * -> vt::numl VT LED
+ * -> vt-numl VT trigger
+ * -> per-device inputx::numl LED
+ *
+ * Userland can however choose the trigger for the vt::numl LED, or
+ * independently choose the trigger for any inputx::numl LED.
+ */
+
+/* VT LED classes and triggers are registered on-demand according to
+ * existing LED devices */
+
+/* Handler for VT LEDs, just triggers the corresponding VT trigger. */
+static void vt_led_set(struct led_classdev *cdev,
+ enum led_brightness brightness);
+static struct led_classdev vt_leds[LED_CNT] = {
+#define DEFINE_INPUT_LED(vt_led, nam, deftrig) \
+ [vt_led] = { \
+ .name = "vt::"nam, \
+ .max_brightness = 1, \
+ .brightness_set = vt_led_set, \
+ .default_trigger = deftrig, \
+ }
+/* Default triggers for the VT LEDs just correspond to the legacy
+ * usage. */
+ DEFINE_INPUT_LED(LED_NUML, "numl", "numlock"),
+ DEFINE_INPUT_LED(LED_CAPSL, "capsl", "capslock"),
+ DEFINE_INPUT_LED(LED_SCROLLL, "scrolll", "scrollock"),
+ DEFINE_INPUT_LED(LED_COMPOSE, "compose", NULL),
+ DEFINE_INPUT_LED(LED_KANA, "kana", "kanalock"),
+ DEFINE_INPUT_LED(LED_SLEEP, "sleep", NULL),
+ DEFINE_INPUT_LED(LED_SUSPEND, "suspend", NULL),
+ DEFINE_INPUT_LED(LED_MUTE, "mute", NULL),
+ DEFINE_INPUT_LED(LED_MISC, "misc", NULL),
+ DEFINE_INPUT_LED(LED_MAIL, "mail", NULL),
+ DEFINE_INPUT_LED(LED_CHARGING, "charging", NULL),
+};
+static const char *const vt_led_names[LED_CNT] = {
+ [LED_NUML] = "numl",
+ [LED_CAPSL] = "capsl",
+ [LED_SCROLLL] = "scrolll",
+ [LED_COMPOSE] = "compose",
+ [LED_KANA] = "kana",
+ [LED_SLEEP] = "sleep",
+ [LED_SUSPEND] = "suspend",
+ [LED_MUTE] = "mute",
+ [LED_MISC] = "misc",
+ [LED_MAIL] = "mail",
+ [LED_CHARGING] = "charging",
+};
+/* Handler for hotplug initialization */
+static void vt_led_trigger_activate(struct led_classdev *cdev);
+/* VT triggers */
+static struct led_trigger vt_led_triggers[LED_CNT] = {
+#define DEFINE_INPUT_LED_TRIGGER(vt_led, nam) \
+ [vt_led] = { \
+ .name = "vt-"nam, \
+ .activate = vt_led_trigger_activate, \
+ }
+ DEFINE_INPUT_LED_TRIGGER(LED_NUML, "numl"),
+ DEFINE_INPUT_LED_TRIGGER(LED_CAPSL, "capsl"),
+ DEFINE_INPUT_LED_TRIGGER(LED_SCROLLL, "scrolll"),
+ DEFINE_INPUT_LED_TRIGGER(LED_COMPOSE, "compose"),
+ DEFINE_INPUT_LED_TRIGGER(LED_KANA, "kana"),
+ DEFINE_INPUT_LED_TRIGGER(LED_SLEEP, "sleep"),
+ DEFINE_INPUT_LED_TRIGGER(LED_SUSPEND, "suspend"),
+ DEFINE_INPUT_LED_TRIGGER(LED_MUTE, "mute"),
+ DEFINE_INPUT_LED_TRIGGER(LED_MISC, "misc"),
+ DEFINE_INPUT_LED_TRIGGER(LED_MAIL, "mail"),
+ DEFINE_INPUT_LED_TRIGGER(LED_CHARGING, "charging"),
+};
+/* Lock for registration coherency */
+static DEFINE_MUTEX(vt_led_registered_lock);
+/* Which VT LED classes and triggers are registered */
+static unsigned long vt_led_registered[BITS_TO_LONGS(LED_CNT)];
+/* Number of input devices having each LED */
+static int vt_led_references[LED_CNT];
+
+/* VT LED state change, tell the VT trigger. */
+static void vt_led_set(struct led_classdev *cdev,
+ enum led_brightness brightness)
+{
+ int led = cdev - vt_leds;
+
+ led_trigger_event(&vt_led_triggers[led], !!brightness);
+}
+
+/* LED state change for some keyboard, notify that keyboard. */
+static void perdevice_input_led_set(struct led_classdev *cdev,
+ enum led_brightness brightness)
+{
+ struct input_dev *dev;
+ struct led_classdev *leds;
+ int led;
+
+ dev = cdev->dev->platform_data;
+ if (!dev)
+ /* Still initializing */
+ return;
+ leds = dev->leds;
+ led = cdev - leds;
+
+ input_event(dev, EV_LED, led, !!brightness);
+ input_event(dev, EV_SYN, SYN_REPORT, 0);
+}
+
+/* Keyboard hotplug, initialize its LED status */
+static void vt_led_trigger_activate(struct led_classdev *cdev)
+{
+ struct led_trigger *trigger = cdev->trigger;
+ int led = trigger - vt_led_triggers;
+
+ if (cdev->brightness_set)
+ cdev->brightness_set(cdev, vt_leds[led].brightness);
+}
+
+/* Free led stuff from input device, used at abortion and disconnection. */
+static void input_led_delete(struct input_dev *dev)
+{
+ if (dev) {
+ struct led_classdev *leds = dev->leds;
+ if (leds) {
+ int i;
+ for (i = 0; i < LED_CNT; i++)
+ kfree(leds[i].name);
+ kfree(leds);
+ dev->leds = NULL;
+ }
+ }
+}
+
+/* A new input device with potential LEDs to connect. */
+int input_led_connect(struct input_dev *dev)
+{
+ int i, error = 0;
+ struct led_classdev *leds;
+
+ dev->leds = leds = kzalloc(sizeof(*leds) * LED_CNT, GFP_KERNEL);
+ if (!dev->leds)
+ return -ENOMEM;
+
+ /* lazily register missing VT LEDs */
+ mutex_lock(&vt_led_registered_lock);
+ for (i = 0; i < LED_CNT; i++)
+ if (vt_leds[i].name && test_bit(i, dev->ledbit)) {
+ if (!vt_led_references[i]) {
+ led_trigger_register(&vt_led_triggers[i]);
+ /* This keyboard is first to have led i,
+ * try to register it */
+ if (!led_classdev_register(NULL, &vt_leds[i]))
+ vt_led_references[i] = 1;
+ else
+ led_trigger_unregister(&vt_led_triggers[i]);
+ } else
+ vt_led_references[i]++;
+ }
+ mutex_unlock(&vt_led_registered_lock);
+
+ /* and register this device's LEDs */
+ for (i = 0; i < LED_CNT; i++)
+ if (vt_leds[i].name && test_bit(i, dev->ledbit)) {
+ leds[i].name = kasprintf(GFP_KERNEL, "%s::%s",
+ dev_name(&dev->dev),
+ vt_led_names[i]);
+ if (!leds[i].name) {
+ error = -ENOMEM;
+ goto err;
+ }
+ leds[i].max_brightness = 1;
+ leds[i].brightness_set = perdevice_input_led_set;
+ leds[i].default_trigger = vt_led_triggers[i].name;
+ }
+
+ /* No issue so far, we can register for real. */
+ for (i = 0; i < LED_CNT; i++)
+ if (leds[i].name) {
+ led_classdev_register(&dev->dev, &leds[i]);
+ leds[i].dev->platform_data = dev;
+ perdevice_input_led_set(&leds[i],
+ vt_leds[i].brightness);
+ }
+
+ return 0;
+
+err:
+ input_led_delete(dev);
+ return error;
+}
+
+/* Disconnected input device. Clean it, and deregister now-useless VT LEDs
+ * and triggers. */
+extern void input_led_disconnect(struct input_dev *dev)
+{
+ int i;
+ struct led_classdev *leds = dev->leds;
+
+ for (i = 0; i < LED_CNT; i++)
+ if (leds[i].name)
+ led_classdev_unregister(&leds[i]);
+
+ input_led_delete(dev);
+
+ mutex_lock(&vt_led_registered_lock);
+ for (i = 0; i < LED_CNT; i++) {
+ if (!vt_leds[i].name || !test_bit(i, dev->ledbit))
+ continue;
+
+ vt_led_references[i]--;
+ if (vt_led_references[i]) {
+ /* Still some devices needing it */
+ continue;
+ }
+
+ led_classdev_unregister(&vt_leds[i]);
+ led_trigger_unregister(&vt_led_triggers[i]);
+ clear_bit(i, vt_led_registered);
+ }
+ mutex_unlock(&vt_led_registered_lock);
+}
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("User LED support for input layer");
+MODULE_AUTHOR("Samuel Thibault <[email protected]>");


2013-07-12 11:36:59

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Hi!

> This permits to reassign keyboard LEDs to something else than keyboard "leds"
> state, by adding keyboard led and modifier triggers connected to a series
> of VT input LEDs, themselves connected to VT input triggers, which
> per-input device LEDs use by default. Userland can thus easily change the LED
> behavior of (a priori) all input devices, or of particular input devices.

Nice! Leds now have proper /sys interface.

But... I boot up, switch from X to console, press capslock, and no
reaction anywhere.

Note that this is notebook with usb keyboard plugged in (and two
monitors), but I believe this worked before...

Thanks,
Pavel

--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

2013-07-12 12:42:50

by Samuel Thibault

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Pavel Machek, le Fri 12 Jul 2013 13:36:56 +0200, a ?crit :
> > This permits to reassign keyboard LEDs to something else than keyboard "leds"
> > state, by adding keyboard led and modifier triggers connected to a series
> > of VT input LEDs, themselves connected to VT input triggers, which
> > per-input device LEDs use by default. Userland can thus easily change the LED
> > behavior of (a priori) all input devices, or of particular input devices.
>
> Nice! Leds now have proper /sys interface.
>
> But... I boot up, switch from X to console, press capslock, and no
> reaction anywhere.

Is it working without the patch? Console-setup for instance is known to
have broken the capslock LED, which is precisely one of the reasons for
this patch, which will provide console-setup with a way to bring back
caps lock working properly.

At any rate, please provide way more information about your keyboard
and LED configuration (output of dumpkeys, dmesg, content of
/sys/class/leds/*/trigger, etc.), as things are just working fine for me
(just like it has been for the past two years).

> Note that this is notebook with usb keyboard plugged in (and two
> monitors), but I believe this worked before...

Things work fine with my USB keyboard too, is this perhaps using an odd
driver which would not expose LEDs in a standard way?

Samuel

2013-07-12 23:33:06

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Hi!

> > > This permits to reassign keyboard LEDs to something else than keyboard "leds"
> > > state, by adding keyboard led and modifier triggers connected to a series
> > > of VT input LEDs, themselves connected to VT input triggers, which
> > > per-input device LEDs use by default. Userland can thus easily change the LED
> > > behavior of (a priori) all input devices, or of particular input devices.
> >
> > Nice! Leds now have proper /sys interface.
> >
> > But... I boot up, switch from X to console, press capslock, and no
> > reaction anywhere.
>
> Is it working without the patch? Console-setup for instance is known to
> have broken the capslock LED, which is precisely one of the reasons for
> this patch, which will provide console-setup with a way to bring back
> caps lock working properly.

You are right, it was broken before.

> At any rate, please provide way more information about your keyboard
> and LED configuration (output of dumpkeys, dmesg, content of
> /sys/class/leds/*/trigger, etc.), as things are just working fine for me
> (just like it has been for the past two years).

Well... I just verified, and it works "as well as before", which it
should.

> > Note that this is notebook with usb keyboard plugged in (and two
> > monitors), but I believe this worked before...
>
> Things work fine with my USB keyboard too, is this perhaps using an odd
> driver which would not expose LEDs in a standard way?

No, everything works as well as it did. Feel free to add:

Tested-by: Pavel Machek <[email protected]>

Thanks,
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

2013-07-13 09:35:58

by Samuel Thibault

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Pavel Machek, le Sat 13 Jul 2013 01:33:01 +0200, a ?crit :
> > > > This permits to reassign keyboard LEDs to something else than keyboard "leds"
> > > > state, by adding keyboard led and modifier triggers connected to a series
> > > > of VT input LEDs, themselves connected to VT input triggers, which
> > > > per-input device LEDs use by default. Userland can thus easily change the LED
> > > > behavior of (a priori) all input devices, or of particular input devices.
> > >
> > > Nice! Leds now have proper /sys interface.
> > >
> > > But... I boot up, switch from X to console, press capslock, and no
> > > reaction anywhere.
> >
> > Is it working without the patch? Console-setup for instance is known to
> > have broken the capslock LED, which is precisely one of the reasons for
> > this patch, which will provide console-setup with a way to bring back
> > caps lock working properly.
>
> You are right, it was broken before.

Ok. You then may want fix your setup by configuring your caps lock led
the way console-setup will be supposed to do in the future:

echo ctrlllock > /sys/class/leds/vt::capsl/trigger

Or some other of the locks that console-setup might be using to
implement caps lock.

> > Things work fine with my USB keyboard too, is this perhaps using an odd
> > driver which would not expose LEDs in a standard way?
>
> No, everything works as well as it did. Feel free to add:
>
> Tested-by: Pavel Machek <[email protected]>

Thanks for testing,
Samuel

2013-07-15 09:12:35

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Hi!

> > > > > This permits to reassign keyboard LEDs to something else than keyboard "leds"
> > > > > state, by adding keyboard led and modifier triggers connected to a series
> > > > > of VT input LEDs, themselves connected to VT input triggers, which
> > > > > per-input device LEDs use by default. Userland can thus easily change the LED
> > > > > behavior of (a priori) all input devices, or of particular input devices.
> > > >
> > > > Nice! Leds now have proper /sys interface.
> > > >
> > > > But... I boot up, switch from X to console, press capslock, and no
> > > > reaction anywhere.
> > >
> > > Is it working without the patch? Console-setup for instance is known to
> > > have broken the capslock LED, which is precisely one of the reasons for
> > > this patch, which will provide console-setup with a way to bring back
> > > caps lock working properly.
> >
> > You are right, it was broken before.
>
> Ok. You then may want fix your setup by configuring your caps lock led
> the way console-setup will be supposed to do in the future:
>
> echo ctrlllock > /sys/class/leds/vt::capsl/trigger

And this indeed makes capslock led work on console, again. Nice!

Richard, could we get this applied for 3.12 or something? It makes
keyboard LEDs sane...

Thanks,
Pavel

--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

2013-07-15 09:27:30

by Peter Korsgaard

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

>>>>> "Samuel" == Samuel Thibault <[email protected]> writes:

Samuel> This permits to reassign keyboard LEDs to something else than keyboard "leds"
Samuel> state, by adding keyboard led and modifier triggers connected to a series
Samuel> of VT input LEDs, themselves connected to VT input triggers, which
Samuel> per-input device LEDs use by default. Userland can thus easily change the LED
Samuel> behavior of (a priori) all input devices, or of particular input devices.

Samuel> This also permits to fix #7063 from userland by using a modifier to implement
Samuel> proper CapsLock behavior and have the keyboard caps lock led show that modifier
Samuel> state.

Samuel> Signed-off-by: Samuel Thibault <[email protected]>
Samuel> [[email protected]: Rebased to 3.2-rc1 or so, cleaned up some includes, and fixed some constants]
Samuel> Signed-off-by: Evan Broder <[email protected]>

Looks good to me. Samuel, thanks for picking this up again.

Acked-by: Peter Korsgaard <[email protected]>

--
Bye, Peter Korsgaard

2013-07-15 15:03:12

by David Herrmann

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Hi

On Sun, Jul 7, 2013 at 12:10 PM, Samuel Thibault
<[email protected]> wrote:
> This permits to reassign keyboard LEDs to something else than keyboard "leds"
> state, by adding keyboard led and modifier triggers connected to a series
> of VT input LEDs, themselves connected to VT input triggers, which
> per-input device LEDs use by default. Userland can thus easily change the LED
> behavior of (a priori) all input devices, or of particular input devices.
>
> This also permits to fix #7063 from userland by using a modifier to implement
> proper CapsLock behavior and have the keyboard caps lock led show that modifier
> state.
>
> Signed-off-by: Samuel Thibault <[email protected]>
> [[email protected]: Rebased to 3.2-rc1 or so, cleaned up some includes, and fixed some constants]
> Signed-off-by: Evan Broder <[email protected]>
>
> diff --exclude .svn -ur linux-3.10-orig/Documentation/leds/leds-class.txt linux-3.10-perso/Documentation/leds/leds-class.txt
> --- linux-3.10-orig/Documentation/leds/leds-class.txt 2013-04-29 13:27:16.959258613 +0200
> +++ linux-3.10-perso/Documentation/leds/leds-class.txt 2013-07-01 23:51:39.229318781 +0200
> @@ -2,9 +2,6 @@
> LED handling under Linux
> ========================
>
> -If you're reading this and thinking about keyboard leds, these are
> -handled by the input subsystem and the led class is *not* needed.
> -
> In its simplest form, the LED class just allows control of LEDs from
> userspace. LEDs appear in /sys/class/leds/. The maximum brightness of the
> LED is defined in max_brightness file. The brightness file will set the brightness
> diff --exclude .svn -ur linux-3.10-orig/drivers/input/input.c linux-3.10-perso/drivers/input/input.c
> --- linux-3.10-orig/drivers/input/input.c 2013-04-29 13:27:16.959258613 +0200
> +++ linux-3.10-perso/drivers/input/input.c 2013-07-01 23:51:39.233318421 +0200
> @@ -28,6 +28,7 @@
> #include <linux/mutex.h>
> #include <linux/rcupdate.h>
> #include "input-compat.h"
> +#include "leds.h"
>
> MODULE_AUTHOR("Vojtech Pavlik <[email protected]>");
> MODULE_DESCRIPTION("Input core");
> @@ -708,6 +709,11 @@
> handle->open = 0;
>
> spin_unlock_irq(&dev->event_lock);
> +
> +#ifdef CONFIG_INPUT_LEDS
> + if (is_event_supported(EV_LED, dev->evbit, EV_MAX))
> + input_led_disconnect(dev);
> +#endif

Please turn "input_led_disconnect()" into a dummy instead of #ifdef
around function calls. See below.

> }
>
> /**
> @@ -2092,6 +2098,11 @@
>
> list_add_tail(&dev->node, &input_dev_list);
>
> +#ifdef CONFIG_INPUT_LEDS
> + if (is_event_supported(EV_LED, dev->evbit, EV_MAX))
> + input_led_connect(dev);
> +#endif
> +

Same as above.

> list_for_each_entry(handler, &input_handler_list, node)
> input_attach_handler(dev, handler);
>
> diff --exclude .svn -ur linux-3.10-orig/drivers/input/Kconfig linux-3.10-perso/drivers/input/Kconfig
> --- linux-3.10-orig/drivers/input/Kconfig 2013-04-29 13:32:29.997497163 +0200
> +++ linux-3.10-perso/drivers/input/Kconfig 2013-07-01 23:51:39.233318421 +0200
> @@ -178,6 +178,15 @@
>
> source "drivers/input/keyboard/Kconfig"
>
> +config INPUT_LEDS
> + tristate "LED Support"
> + depends on LEDS_CLASS
> + select LEDS_TRIGGERS
> + default y
> + help
> + This option enables support for the LEDs on keyboard managed

-the +keyboard_s_?

> + by the input layer.
> +
> source "drivers/input/mouse/Kconfig"
>
> source "drivers/input/joystick/Kconfig"
> diff --exclude .svn -ur linux-3.10-orig/drivers/input/Makefile linux-3.10-perso/drivers/input/Makefile
> --- linux-3.10-orig/drivers/input/Makefile 2013-04-29 13:27:16.959258613 +0200
> +++ linux-3.10-perso/drivers/input/Makefile 2013-07-01 23:51:39.233318421 +0200
> @@ -18,6 +18,7 @@
> obj-$(CONFIG_INPUT_EVBUG) += evbug.o
>
> obj-$(CONFIG_INPUT_KEYBOARD) += keyboard/
> +obj-$(CONFIG_INPUT_LEDS) += leds.o

Nitpick, but could you move it into the section above or below? This
section includes sub-directories only.

> obj-$(CONFIG_INPUT_MOUSE) += mouse/
> obj-$(CONFIG_INPUT_JOYSTICK) += joystick/
> obj-$(CONFIG_INPUT_TABLET) += tablet/
> diff --exclude .svn -ur linux-3.10-orig/drivers/leds/Kconfig linux-3.10-perso/drivers/leds/Kconfig
> --- linux-3.10-orig/drivers/leds/Kconfig 2013-07-01 01:56:00.335874214 +0200
> +++ linux-3.10-perso/drivers/leds/Kconfig 2013-07-01 23:51:39.233318421 +0200
> @@ -11,9 +11,6 @@
> Say Y to enable Linux LED support. This allows control of supported
> LEDs from both userspace and optionally, by kernel events (triggers).
>
> - This is not related to standard keyboard LEDs which are controlled
> - via the input system.
> -
> if NEW_LEDS
>
> config LEDS_CLASS
> diff --exclude .svn -ur linux-3.10-orig/drivers/tty/Kconfig linux-3.10-perso/drivers/tty/Kconfig
> --- linux-3.10-orig/drivers/tty/Kconfig 2013-04-29 13:32:36.353298705 +0200
> +++ linux-3.10-perso/drivers/tty/Kconfig 2013-07-01 23:51:39.237318056 +0200
> @@ -13,6 +13,10 @@
> bool "Virtual terminal" if EXPERT
> depends on !S390 && !UML
> select INPUT
> + select NEW_LEDS
> + select LEDS_CLASS
> + select LEDS_TRIGGERS
> + select INPUT_LEDS

This looks odd. Any dependency changes in the input-layer will break
this. But I guess that's what we get for a non-recursive "select". Hmm

> default y
> ---help---
> If you say Y here, you will get support for terminal devices with
> diff --exclude .svn -ur linux-3.10-orig/drivers/tty/vt/keyboard.c linux-3.10-perso/drivers/tty/vt/keyboard.c
> --- linux-3.10-orig/drivers/tty/vt/keyboard.c 2013-04-29 13:32:36.681288463 +0200
> +++ linux-3.10-perso/drivers/tty/vt/keyboard.c 2013-07-01 23:51:39.237318056 +0200
> @@ -33,6 +33,7 @@
> #include <linux/string.h>
> #include <linux/init.h>
> #include <linux/slab.h>
> +#include <linux/leds.h>
>
> #include <linux/kbd_kern.h>
> #include <linux/kbd_diacr.h>
> @@ -130,6 +131,7 @@
> static int shift_state = 0;
>
> static unsigned char ledstate = 0xff; /* undefined */
> +static unsigned char lockstate = 0xff; /* undefined */
> static unsigned char ledioctl;
>
> static struct ledptr {
> @@ -967,6 +969,32 @@
> }
> }
>
> +/* We route VT keyboard "leds" through triggers */
> +static void kbd_ledstate_trigger_activate(struct led_classdev *cdev);
> +static struct led_trigger ledtrig_ledstate[] = {
> +#define DEFINE_LEDSTATE_TRIGGER(kbd_led, nam) \
> + [kbd_led] = { .name = nam, .activate = kbd_ledstate_trigger_activate, }
> + DEFINE_LEDSTATE_TRIGGER(VC_SCROLLOCK, "scrollock"),
> + DEFINE_LEDSTATE_TRIGGER(VC_NUMLOCK, "numlock"),
> + DEFINE_LEDSTATE_TRIGGER(VC_CAPSLOCK, "capslock"),
> + DEFINE_LEDSTATE_TRIGGER(VC_KANALOCK, "kanalock"),
> +#undef DEFINE_LEDSTATE_TRIGGER
> +};
> +static void kbd_lockstate_trigger_activate(struct led_classdev *cdev);
> +static struct led_trigger ledtrig_lockstate[] = {
> +#define DEFINE_LOCKSTATE_TRIGGER(kbd_led, nam) \
> + [kbd_led] = { .name = nam, .activate = kbd_lockstate_trigger_activate, }
> + DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLOCK, "shiftlock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_ALTGRLOCK, "altgrlock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_CTRLLOCK, "ctrllock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_ALTLOCK, "altlock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLLOCK, "shiftllock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTRLOCK, "shiftrlock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_CTRLLLOCK, "ctrlllock"),
> + DEFINE_LOCKSTATE_TRIGGER(VC_CTRLRLOCK, "ctrlrlock"),
> +#undef DEFINE_LOCKSTATE_TRIGGER
> +};
> +

We use empty lines between function-declarations and variable
declarations, I think. Would make this a lot easier to read. I also
think that the macros don't really simplify this. So:
[VC_SHIFTLOCK] = { .name = "shiftlock", .activate =
kbd_lockstate_trigger_activate },
isn't much longer than:
DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLOCK, "shiftlock"),
But I guess that's a matter of taste.

> /*
> * The leds display either (i) the status of NumLock, CapsLock, ScrollLock,
> * or (ii) whatever pattern of lights people want to show using KDSETLED,
> @@ -1002,6 +1030,7 @@
>
> leds = kbd->ledflagstate;
>
> + /* TODO: should be replaced by a LED trigger */
> if (kbd->ledmode == LED_SHOW_MEM) {
> for (i = 0; i < 3; i++)
> if (ledptrs[i].valid) {
> @@ -1014,18 +1043,24 @@
> return leds;
> }
>
> -static int kbd_update_leds_helper(struct input_handle *handle, void *data)
> +/* Called on trigger connection, to set initial state */
> +static void kbd_ledstate_trigger_activate(struct led_classdev *cdev)
> {
> - unsigned char leds = *(unsigned char *)data;
> + struct led_trigger *trigger = cdev->trigger;
> + int led = trigger - ledtrig_ledstate;
>
> - if (test_bit(EV_LED, handle->dev->evbit)) {
> - input_inject_event(handle, EV_LED, LED_SCROLLL, !!(leds & 0x01));
> - input_inject_event(handle, EV_LED, LED_NUML, !!(leds & 0x02));
> - input_inject_event(handle, EV_LED, LED_CAPSL, !!(leds & 0x04));
> - input_inject_event(handle, EV_SYN, SYN_REPORT, 0);
> - }
> + tasklet_disable(&keyboard_tasklet);
> + led_trigger_event(trigger, ledstate & (1 << led) ? LED_FULL : LED_OFF);
> + tasklet_enable(&keyboard_tasklet);
> +}
> +static void kbd_lockstate_trigger_activate(struct led_classdev *cdev)
> +{
> + struct led_trigger *trigger = cdev->trigger;
> + int led = trigger - ledtrig_lockstate;
>
> - return 0;
> + tasklet_disable(&keyboard_tasklet);
> + led_trigger_event(trigger, lockstate & (1 << led) ? LED_FULL : LED_OFF);
> + tasklet_enable(&keyboard_tasklet);
> }

Again, please add empty lines between functions definitions.

>
> /**
> @@ -1120,10 +1155,24 @@
> spin_unlock_irqrestore(&led_lock, flags);
>
> if (leds != ledstate) {
> - input_handler_for_each_handle(&kbd_handler, &leds,
> - kbd_update_leds_helper);
> + int i;

Move "int i;" to the top. You use it again below. gcc is smart enough
to notice that both usages are independent.

> + for (i = 0; i < ARRAY_SIZE(ledtrig_ledstate); i++)
> + if ((leds ^ ledstate) & (1 << i))
> + led_trigger_event(&ledtrig_ledstate[i],
> + leds & (1 << i)
> + ? LED_FULL : LED_OFF);
> ledstate = leds;
> }
> +
> + if (kbd->lockstate != lockstate) {
> + int i;
> + for (i = 0; i < ARRAY_SIZE(ledtrig_lockstate); i++)
> + if ((kbd->lockstate ^ lockstate) & (1 << i))
> + led_trigger_event(&ledtrig_lockstate[i],
> + kbd->lockstate & (1 << i)
> + ? LED_FULL : LED_OFF);
> + lockstate = kbd->lockstate;
> + }
> }
>
> DECLARE_TASKLET_DISABLED(keyboard_tasklet, kbd_bh, 0);
> @@ -1461,20 +1510,6 @@
> kfree(handle);
> }
>
> -/*
> - * Start keyboard handler on the new keyboard by refreshing LED state to
> - * match the rest of the system.
> - */
> -static void kbd_start(struct input_handle *handle)
> -{
> - tasklet_disable(&keyboard_tasklet);
> -
> - if (ledstate != 0xff)
> - kbd_update_leds_helper(handle, &ledstate);
> -
> - tasklet_enable(&keyboard_tasklet);
> -}
> -
> static const struct input_device_id kbd_ids[] = {
> {
> .flags = INPUT_DEVICE_ID_MATCH_EVBIT,
> @@ -1496,7 +1531,6 @@
> .match = kbd_match,
> .connect = kbd_connect,
> .disconnect = kbd_disconnect,
> - .start = kbd_start,
> .name = "kbd",
> .id_table = kbd_ids,
> };
> @@ -1520,6 +1554,11 @@
> if (error)
> return error;
>
> + for (i = 0; i < ARRAY_SIZE(ledtrig_ledstate); i++)
> + led_trigger_register(&ledtrig_ledstate[i]);
> + for (i = 0; i < ARRAY_SIZE(ledtrig_lockstate); i++)
> + led_trigger_register(&ledtrig_lockstate[i]);
> +

This returns "int", are you sure no error-handling is needed?

> tasklet_enable(&keyboard_tasklet);
> tasklet_schedule(&keyboard_tasklet);
>
> diff --exclude .svn -ur linux-3.10-orig/include/linux/input.h linux-3.10-perso/include/linux/input.h
> --- linux-3.10-orig/include/linux/input.h 2013-04-29 13:27:16.959258613 +0200
> +++ linux-3.10-perso/include/linux/input.h 2013-07-01 23:51:39.241317685 +0200
> @@ -164,6 +164,8 @@
> unsigned long snd[BITS_TO_LONGS(SND_CNT)];
> unsigned long sw[BITS_TO_LONGS(SW_CNT)];
>
> + struct led_classdev *leds;
> +
> int (*open)(struct input_dev *dev);
> void (*close)(struct input_dev *dev);
> int (*flush)(struct input_dev *dev, struct file *file);
> --- linux-3.10-orig/drivers/input/leds.h 1970-01-01 01:00:00.000000000 +0100
> +++ linux-3.10-perso/drivers/input/leds.h 2011-11-14 02:24:26.738456456 +0100
> @@ -0,0 +1,14 @@
> +/*
> + * LED support for the input layer
> + *
> + * Copyright 2010-2013 Samuel Thibault <[email protected]>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + */
> +
> +#include <linux/input.h>
> +
> +extern int input_led_connect(struct input_dev *dev);
> +extern void input_led_disconnect(struct input_dev *dev);

Header-protection is missing. Besides, these 2 lines could be easily
merged into input.h. Any reason not to?
Also remove the "extern" before functions. It does not add any value
and we try to drop it for new declarations.

And you can avoid the #ifdef mentioned above if you use something like this:

#ifdef CONFIG_LEDS

int input_led_connect(struct input_dev *dev);
void input_led_disconnect(struct input_dev *dev);

#else

static inline int input_led_connect(struct input_dev *dev)
{
return 0;
}

static inline void input_led_disconnect(struct input_dev *dev)
{
}

#endif

> --- linux-3.10-orig/drivers/input/leds.c 1970-01-01 01:00:00.000000000 +0100
> +++ linux-3.10-perso/drivers/input/leds.c 2011-11-14 03:23:07.578469395 +0100
> @@ -0,0 +1,243 @@
> +/*
> + * LED support for the input layer
> + *
> + * Copyright 2010-2013 Samuel Thibault <[email protected]>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + */
> +
> +#include <linux/kernel.h>
> +#include <linux/slab.h>
> +#include <linux/module.h>
> +#include <linux/init.h>
> +#include <linux/leds.h>
> +#include <linux/input.h>

If you keep "input/leds.h" you should probably include it here. If you
drop it, you're fine.

> +
> +/*
> + * Keyboard LEDs are propagated by default like the following example:
> + *
> + * VT keyboard numlock trigger
> + * -> vt::numl VT LED
> + * -> vt-numl VT trigger

Nitpick: What's the reason for the syntax difference? "vt-numl" vs. "vt::numl"

> + * -> per-device inputx::numl LED

nitpick, but upper-case "X" is probably easier to understand. I had to
read it twice to get what the "x" means.

> + *
> + * Userland can however choose the trigger for the vt::numl LED, or
> + * independently choose the trigger for any inputx::numl LED.
> + */
> +
> +/* VT LED classes and triggers are registered on-demand according to
> + * existing LED devices */

Merge this comment with the comment above?

> +
> +/* Handler for VT LEDs, just triggers the corresponding VT trigger. */
> +static void vt_led_set(struct led_classdev *cdev,
> + enum led_brightness brightness);
> +static struct led_classdev vt_leds[LED_CNT] = {
> +#define DEFINE_INPUT_LED(vt_led, nam, deftrig) \
> + [vt_led] = { \
> + .name = "vt::"nam, \
> + .max_brightness = 1, \
> + .brightness_set = vt_led_set, \
> + .default_trigger = deftrig, \
> + }
> +/* Default triggers for the VT LEDs just correspond to the legacy
> + * usage. */
> + DEFINE_INPUT_LED(LED_NUML, "numl", "numlock"),
> + DEFINE_INPUT_LED(LED_CAPSL, "capsl", "capslock"),
> + DEFINE_INPUT_LED(LED_SCROLLL, "scrolll", "scrollock"),
> + DEFINE_INPUT_LED(LED_COMPOSE, "compose", NULL),
> + DEFINE_INPUT_LED(LED_KANA, "kana", "kanalock"),
> + DEFINE_INPUT_LED(LED_SLEEP, "sleep", NULL),
> + DEFINE_INPUT_LED(LED_SUSPEND, "suspend", NULL),
> + DEFINE_INPUT_LED(LED_MUTE, "mute", NULL),
> + DEFINE_INPUT_LED(LED_MISC, "misc", NULL),
> + DEFINE_INPUT_LED(LED_MAIL, "mail", NULL),
> + DEFINE_INPUT_LED(LED_CHARGING, "charging", NULL),
> +};
> +static const char *const vt_led_names[LED_CNT] = {
> + [LED_NUML] = "numl",
> + [LED_CAPSL] = "capsl",
> + [LED_SCROLLL] = "scrolll",
> + [LED_COMPOSE] = "compose",
> + [LED_KANA] = "kana",
> + [LED_SLEEP] = "sleep",
> + [LED_SUSPEND] = "suspend",
> + [LED_MUTE] = "mute",
> + [LED_MISC] = "misc",
> + [LED_MAIL] = "mail",
> + [LED_CHARGING] = "charging",
> +};
> +/* Handler for hotplug initialization */
> +static void vt_led_trigger_activate(struct led_classdev *cdev);
> +/* VT triggers */
> +static struct led_trigger vt_led_triggers[LED_CNT] = {
> +#define DEFINE_INPUT_LED_TRIGGER(vt_led, nam) \
> + [vt_led] = { \
> + .name = "vt-"nam, \
> + .activate = vt_led_trigger_activate, \
> + }
> + DEFINE_INPUT_LED_TRIGGER(LED_NUML, "numl"),
> + DEFINE_INPUT_LED_TRIGGER(LED_CAPSL, "capsl"),
> + DEFINE_INPUT_LED_TRIGGER(LED_SCROLLL, "scrolll"),
> + DEFINE_INPUT_LED_TRIGGER(LED_COMPOSE, "compose"),
> + DEFINE_INPUT_LED_TRIGGER(LED_KANA, "kana"),
> + DEFINE_INPUT_LED_TRIGGER(LED_SLEEP, "sleep"),
> + DEFINE_INPUT_LED_TRIGGER(LED_SUSPEND, "suspend"),
> + DEFINE_INPUT_LED_TRIGGER(LED_MUTE, "mute"),
> + DEFINE_INPUT_LED_TRIGGER(LED_MISC, "misc"),
> + DEFINE_INPUT_LED_TRIGGER(LED_MAIL, "mail"),
> + DEFINE_INPUT_LED_TRIGGER(LED_CHARGING, "charging"),
> +};
> +/* Lock for registration coherency */
> +static DEFINE_MUTEX(vt_led_registered_lock);
> +/* Which VT LED classes and triggers are registered */
> +static unsigned long vt_led_registered[BITS_TO_LONGS(LED_CNT)];
> +/* Number of input devices having each LED */
> +static int vt_led_references[LED_CNT];

Please add some empty lines here. At least the different static arrays
should be separated by an empty newline.

> +
> +/* VT LED state change, tell the VT trigger. */
> +static void vt_led_set(struct led_classdev *cdev,
> + enum led_brightness brightness)
> +{
> + int led = cdev - vt_leds;
> +
> + led_trigger_event(&vt_led_triggers[led], !!brightness);
> +}
> +
> +/* LED state change for some keyboard, notify that keyboard. */
> +static void perdevice_input_led_set(struct led_classdev *cdev,
> + enum led_brightness brightness)
> +{
> + struct input_dev *dev;
> + struct led_classdev *leds;
> + int led;
> +
> + dev = cdev->dev->platform_data;
> + if (!dev)
> + /* Still initializing */
> + return;
> + leds = dev->leds;
> + led = cdev - leds;
> +
> + input_event(dev, EV_LED, led, !!brightness);
> + input_event(dev, EV_SYN, SYN_REPORT, 0);
> +}
> +
> +/* Keyboard hotplug, initialize its LED status */
> +static void vt_led_trigger_activate(struct led_classdev *cdev)
> +{
> + struct led_trigger *trigger = cdev->trigger;
> + int led = trigger - vt_led_triggers;
> +
> + if (cdev->brightness_set)
> + cdev->brightness_set(cdev, vt_leds[led].brightness);
> +}
> +
> +/* Free led stuff from input device, used at abortion and disconnection. */
> +static void input_led_delete(struct input_dev *dev)
> +{
> + if (dev) {
> + struct led_classdev *leds = dev->leds;
> + if (leds) {
> + int i;
> + for (i = 0; i < LED_CNT; i++)
> + kfree(leds[i].name);
> + kfree(leds);
> + dev->leds = NULL;
> + }
> + }
> +}
> +
> +/* A new input device with potential LEDs to connect. */
> +int input_led_connect(struct input_dev *dev)
> +{
> + int i, error = 0;
> + struct led_classdev *leds;
> +
> + dev->leds = leds = kzalloc(sizeof(*leds) * LED_CNT, GFP_KERNEL);
> + if (!dev->leds)
> + return -ENOMEM;
> +
> + /* lazily register missing VT LEDs */
> + mutex_lock(&vt_led_registered_lock);
> + for (i = 0; i < LED_CNT; i++)
> + if (vt_leds[i].name && test_bit(i, dev->ledbit)) {
> + if (!vt_led_references[i]) {
> + led_trigger_register(&vt_led_triggers[i]);
> + /* This keyboard is first to have led i,
> + * try to register it */
> + if (!led_classdev_register(NULL, &vt_leds[i]))
> + vt_led_references[i] = 1;
> + else
> + led_trigger_unregister(&vt_led_triggers[i]);
> + } else
> + vt_led_references[i]++;
> + }
> + mutex_unlock(&vt_led_registered_lock);
> +
> + /* and register this device's LEDs */
> + for (i = 0; i < LED_CNT; i++)
> + if (vt_leds[i].name && test_bit(i, dev->ledbit)) {
> + leds[i].name = kasprintf(GFP_KERNEL, "%s::%s",
> + dev_name(&dev->dev),
> + vt_led_names[i]);
> + if (!leds[i].name) {
> + error = -ENOMEM;
> + goto err;
> + }
> + leds[i].max_brightness = 1;
> + leds[i].brightness_set = perdevice_input_led_set;
> + leds[i].default_trigger = vt_led_triggers[i].name;
> + }
> +
> + /* No issue so far, we can register for real. */
> + for (i = 0; i < LED_CNT; i++)
> + if (leds[i].name) {
> + led_classdev_register(&dev->dev, &leds[i]);
> + leds[i].dev->platform_data = dev;
> + perdevice_input_led_set(&leds[i],
> + vt_leds[i].brightness);
> + }
> +
> + return 0;
> +
> +err:
> + input_led_delete(dev);
> + return error;
> +}
> +
> +/* Disconnected input device. Clean it, and deregister now-useless VT LEDs
> + * and triggers. */
> +extern void input_led_disconnect(struct input_dev *dev)
> +{
> + int i;
> + struct led_classdev *leds = dev->leds;
> +
> + for (i = 0; i < LED_CNT; i++)
> + if (leds[i].name)
> + led_classdev_unregister(&leds[i]);
> +
> + input_led_delete(dev);
> +
> + mutex_lock(&vt_led_registered_lock);
> + for (i = 0; i < LED_CNT; i++) {
> + if (!vt_leds[i].name || !test_bit(i, dev->ledbit))
> + continue;
> +
> + vt_led_references[i]--;
> + if (vt_led_references[i]) {
> + /* Still some devices needing it */
> + continue;
> + }
> +
> + led_classdev_unregister(&vt_leds[i]);
> + led_trigger_unregister(&vt_led_triggers[i]);
> + clear_bit(i, vt_led_registered);
> + }
> + mutex_unlock(&vt_led_registered_lock);
> +}
> +
> +MODULE_LICENSE("GPL");
> +MODULE_DESCRIPTION("User LED support for input layer");
> +MODULE_AUTHOR("Samuel Thibault <[email protected]>");

I haven't done any locking review, yet, and some of the comments are
probably just a matter of taste. But the patch looks good.

Cheers
David

> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/

2013-07-15 19:08:23

by Samuel Thibault

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Hello,

David Herrmann, le Mon 15 Jul 2013 17:03:08 +0200, a ?crit :
> > @@ -13,6 +13,10 @@
> > bool "Virtual terminal" if EXPERT
> > depends on !S390 && !UML
> > select INPUT
> > + select NEW_LEDS
> > + select LEDS_CLASS
> > + select LEDS_TRIGGERS
> > + select INPUT_LEDS
>
> This looks odd. Any dependency changes in the input-layer will break
> this. But I guess that's what we get for a non-recursive "select". Hmm

Yes, that's the issue.

> I also think that the macros don't really simplify this. So:
> [VC_SHIFTLOCK] = { .name = "shiftlock", .activate =
> kbd_lockstate_trigger_activate },
> isn't much longer than:
> DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLOCK, "shiftlock"),

That makes it more than 80 columns, at least.

> > + for (i = 0; i < ARRAY_SIZE(ledtrig_ledstate); i++)
> > + led_trigger_register(&ledtrig_ledstate[i]);
> > + for (i = 0; i < ARRAY_SIZE(ledtrig_lockstate); i++)
> > + led_trigger_register(&ledtrig_lockstate[i]);
> > +
>
> This returns "int", are you sure no error-handling is needed?

ATM only registering the same name several times can happen, and
shouldn't ever happen. But better warn than be sorry, indeed.

> > + * Keyboard LEDs are propagated by default like the following example:
> > + *
> > + * VT keyboard numlock trigger
> > + * -> vt::numl VT LED
> > + * -> vt-numl VT trigger
>
> Nitpick: What's the reason for the syntax difference? "vt-numl" vs. "vt::numl"

vt::numl is a clear LED classification convention. For triggers, I don't
see a clear convention, but at least I have never seen "::", only "-".
That makes me realize that for keyboard triggers I haven't introduced a
prefix, and only used "scrollock", "numlock", etc. Perhaps
"kbd-scrollock" etc. or (in phase with LEDs), "kbd::scrollock" etc.?

Samuel

2013-07-17 15:14:15

by David Herrmann

[permalink] [raw]
Subject: Re: [PATCH] Route kbd LEDs through the generic LEDs layer

Hi

On Mon, Jul 15, 2013 at 9:08 PM, Samuel Thibault
<[email protected]> wrote:
> Hello,
>
> David Herrmann, le Mon 15 Jul 2013 17:03:08 +0200, a ?crit :
>> > @@ -13,6 +13,10 @@
>> > bool "Virtual terminal" if EXPERT
>> > depends on !S390 && !UML
>> > select INPUT
>> > + select NEW_LEDS
>> > + select LEDS_CLASS
>> > + select LEDS_TRIGGERS
>> > + select INPUT_LEDS
>>
>> This looks odd. Any dependency changes in the input-layer will break
>> this. But I guess that's what we get for a non-recursive "select". Hmm
>
> Yes, that's the issue.
>
>> I also think that the macros don't really simplify this. So:
>> [VC_SHIFTLOCK] = { .name = "shiftlock", .activate =
>> kbd_lockstate_trigger_activate },
>> isn't much longer than:
>> DEFINE_LOCKSTATE_TRIGGER(VC_SHIFTLOCK, "shiftlock"),
>
> That makes it more than 80 columns, at least.

Indeed, I missed that.

>> > + for (i = 0; i < ARRAY_SIZE(ledtrig_ledstate); i++)
>> > + led_trigger_register(&ledtrig_ledstate[i]);
>> > + for (i = 0; i < ARRAY_SIZE(ledtrig_lockstate); i++)
>> > + led_trigger_register(&ledtrig_lockstate[i]);
>> > +
>>
>> This returns "int", are you sure no error-handling is needed?
>
> ATM only registering the same name several times can happen, and
> shouldn't ever happen. But better warn than be sorry, indeed.

Yepp, it also avoids ignoring any future errors that we might
introduce in led_trigger_register().

>> > + * Keyboard LEDs are propagated by default like the following example:
>> > + *
>> > + * VT keyboard numlock trigger
>> > + * -> vt::numl VT LED
>> > + * -> vt-numl VT trigger
>>
>> Nitpick: What's the reason for the syntax difference? "vt-numl" vs. "vt::numl"
>
> vt::numl is a clear LED classification convention. For triggers, I don't
> see a clear convention, but at least I have never seen "::", only "-".
> That makes me realize that for keyboard triggers I haven't introduced a
> prefix, and only used "scrollock", "numlock", etc. Perhaps
> "kbd-scrollock" etc. or (in phase with LEDs), "kbd::scrollock" etc.?

I understand. Due to lack of insight, I cannot really comment on that,
sorry. Just wanted to go sure you're aware of that. So feel free to
pick any of them.

Thanks
David