1 ===============================
2 Creating an input device driver
3 ===============================
4
5 The simplest example
6 ~~~~~~~~~~~~~~~~~~~~
7
8 Here comes a very simple example of an input device driver. The device has
9 just one button and the button is accessible at i/o port BUTTON_PORT. When
10 pressed or released a BUTTON_IRQ happens. The driver could look like::
11
12 #include <linux/input.h>
13 #include <linux/module.h>
14 #include <linux/init.h>
15
16 #include <asm/irq.h>
17 #include <asm/io.h>
18
19 static struct input_dev *button_dev;
20
21 static irqreturn_t button_interrupt(int irq, void *dummy)
22 {
23 input_report_key(button_dev, BTN_0, inb(BUTTON_PORT) & 1);
24 input_sync(button_dev);
25 return IRQ_HANDLED;
26 }
27
28 static int __init button_init(void)
29 {
30 int error;
31
32 if (request_irq(BUTTON_IRQ, button_interrupt, 0, "button", NULL)) {
33 printk(KERN_ERR "button.c: Can't allocate irq %d\n", button_irq);
34 return -EBUSY;
35 }
36
37 button_dev = input_allocate_device();
38 if (!button_dev) {
39 printk(KERN_ERR "button.c: Not enough memory\n");
40 error = -ENOMEM;
41 goto err_free_irq;
42 }
43
44 button_dev->evbit[0] = BIT_MASK(EV_KEY);
45 button_dev->keybit[BIT_WORD(BTN_0)] = BIT_MASK(BTN_0);
46
47 error = input_register_device(button_dev);
48 if (error) {
49 printk(KERN_ERR "button.c: Failed to register device\n");
50 goto err_free_dev;
51 }
52
53 return 0;
54
55 err_free_dev:
56 input_free_device(button_dev);
57 err_free_irq:
58 free_irq(BUTTON_IRQ, button_interrupt);
59 return error;
60 }
61
62 static void __exit button_exit(void)
63 {
64 input_unregister_device(button_dev);
65 free_irq(BUTTON_IRQ, button_interrupt);
66 }
67
68 module_init(button_init);
69 module_exit(button_exit);
70
71 What the example does
72 ~~~~~~~~~~~~~~~~~~~~~
73
74 First it has to include the <linux/input.h> file, which interfaces to the
75 input subsystem. This provides all the definitions needed.
76
77 In the _init function, which is called either upon module load or when
78 booting the kernel, it grabs the required resources (it should also check
79 for the presence of the device).
80
81 Then it allocates a new input device structure with input_allocate_device()
82 and sets up input bitfields. This way the device driver tells the other
83 parts of the input systems what it is - what events can be generated or
84 accepted by this input device. Our example device can only generate EV_KEY
85 type events, and from those only BTN_0 event code. Thus we only set these
86 two bits. We could have used::
87
88 set_bit(EV_KEY, button_dev->evbit);
89 set_bit(BTN_0, button_dev->keybit);
90
91 as well, but with more than single bits the first approach tends to be
92 shorter.
93
94 Then the example driver registers the input device structure by calling::
95
96 input_register_device(button_dev);
97
98 This adds the button_dev structure to linked lists of the input driver and
99 calls device handler modules _connect functions to tell them a new input
100 device has appeared. input_register_device() may sleep and therefore must
101 not be called from an interrupt or with a spinlock held.
102
103 While in use, the only used function of the driver is::
104
105 button_interrupt()
106
107 which upon every interrupt from the button checks its state and reports it
108 via the::
109
110 input_report_key()
111
112 call to the input system. There is no need to check whether the interrupt
113 routine isn't reporting two same value events (press, press for example) to
114 the input system, because the input_report_* functions check that
115 themselves.
116
117 Then there is the::
118
119 input_sync()
120
121 call to tell those who receive the events that we've sent a complete report.
122 This doesn't seem important in the one button case, but is quite important
123 for example for mouse movement, where you don't want the X and Y values
124 to be interpreted separately, because that'd result in a different movement.
125
126 dev->open() and dev->close()
127 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
128
129 In case the driver has to repeatedly poll the device, because it doesn't
130 have an interrupt coming from it and the polling is too expensive to be done
131 all the time, or if the device uses a valuable resource (e.g. interrupt), it
132 can use the open and close callback to know when it can stop polling or
133 release the interrupt and when it must resume polling or grab the interrupt
134 again. To do that, we would add this to our example driver::
135
136 static int button_open(struct input_dev *dev)
137 {
138 if (request_irq(BUTTON_IRQ, button_interrupt, 0, "button", NULL)) {
139 printk(KERN_ERR "button.c: Can't allocate irq %d\n", button_irq);
140 return -EBUSY;
141 }
142
143 return 0;
144 }
145
146 static void button_close(struct input_dev *dev)
147 {
148 free_irq(IRQ_AMIGA_VERTB, button_interrupt);
149 }
150
151 static int __init button_init(void)
152 {
153 ...
154 button_dev->open = button_open;
155 button_dev->close = button_close;
156 ...
157 }
158
159 Note that input core keeps track of number of users for the device and
160 makes sure that dev->open() is called only when the first user connects
161 to the device and that dev->close() is called when the very last user
162 disconnects. Calls to both callbacks are serialized.
163
164 The open() callback should return a 0 in case of success or any non-zero value
165 in case of failure. The close() callback (which is void) must always succeed.
166
167 Inhibiting input devices
168 ~~~~~~~~~~~~~~~~~~~~~~~~
169
170 Inhibiting a device means ignoring input events from it. As such it is about
171 maintaining relationships with input handlers - either already existing
172 relationships, or relationships to be established while the device is in
173 inhibited state.
174
175 If a device is inhibited, no input handler will receive events from it.
176
177 The fact that nobody wants events from the device is exploited further, by
178 calling device's close() (if there are users) and open() (if there are users) on
179 inhibit and uninhibit operations, respectively. Indeed, the meaning of close()
180 is to stop providing events to the input core and that of open() is to start
181 providing events to the input core.
182
183 Calling the device's close() method on inhibit (if there are users) allows the
184 driver to save power. Either by directly powering down the device or by
185 releasing the runtime-PM reference it got in open() when the driver is using
186 runtime-PM.
187
188 Inhibiting and uninhibiting are orthogonal to opening and closing the device by
189 input handlers. Userspace might want to inhibit a device in anticipation before
190 any handler is positively matched against it.
191
192 Inhibiting and uninhibiting are orthogonal to device's being a wakeup source,
193 too. Being a wakeup source plays a role when the system is sleeping, not when
194 the system is operating. How drivers should program their interaction between
195 inhibiting, sleeping and being a wakeup source is driver-specific.
196
197 Taking the analogy with the network devices - bringing a network interface down
198 doesn't mean that it should be impossible be wake the system up on LAN through
199 this interface. So, there may be input drivers which should be considered wakeup
200 sources even when inhibited. Actually, in many I2C input devices their interrupt
201 is declared a wakeup interrupt and its handling happens in driver's core, which
202 is not aware of input-specific inhibit (nor should it be). Composite devices
203 containing several interfaces can be inhibited on a per-interface basis and e.g.
204 inhibiting one interface shouldn't affect the device's capability of being a
205 wakeup source.
206
207 If a device is to be considered a wakeup source while inhibited, special care
208 must be taken when programming its suspend(), as it might need to call device's
209 open(). Depending on what close() means for the device in question, not
210 opening() it before going to sleep might make it impossible to provide any
211 wakeup events. The device is going to sleep anyway.
212
213 Basic event types
214 ~~~~~~~~~~~~~~~~~
215
216 The most simple event type is EV_KEY, which is used for keys and buttons.
217 It's reported to the input system via::
218
219 input_report_key(struct input_dev *dev, int code, int value)
220
221 See uapi/linux/input-event-codes.h for the allowable values of code (from 0 to
222 KEY_MAX). Value is interpreted as a truth value, i.e. any non-zero value means
223 key pressed, zero value means key released. The input code generates events only
224 in case the value is different from before.
225
226 In addition to EV_KEY, there are two more basic event types: EV_REL and
227 EV_ABS. They are used for relative and absolute values supplied by the
228 device. A relative value may be for example a mouse movement in the X axis.
229 The mouse reports it as a relative difference from the last position,
230 because it doesn't have any absolute coordinate system to work in. Absolute
231 events are namely for joysticks and digitizers - devices that do work in an
232 absolute coordinate systems.
233
234 Having the device report EV_REL buttons is as simple as with EV_KEY; simply
235 set the corresponding bits and call the::
236
237 input_report_rel(struct input_dev *dev, int code, int value)
238
239 function. Events are generated only for non-zero values.
240
241 However EV_ABS requires a little special care. Before calling
242 input_register_device, you have to fill additional fields in the input_dev
243 struct for each absolute axis your device has. If our button device had also
244 the ABS_X axis::
245
246 button_dev.absmin[ABS_X] = 0;
247 button_dev.absmax[ABS_X] = 255;
248 button_dev.absfuzz[ABS_X] = 4;
249 button_dev.absflat[ABS_X] = 8;
250
251 Or, you can just say::
252
253 input_set_abs_params(button_dev, ABS_X, 0, 255, 4, 8);
254
255 This setting would be appropriate for a joystick X axis, with the minimum of
256 0, maximum of 255 (which the joystick *must* be able to reach, no problem if
257 it sometimes reports more, but it must be able to always reach the min and
258 max values), with noise in the data up to +- 4, and with a center flat
259 position of size 8.
260
261 If you don't need absfuzz and absflat, you can set them to zero, which mean
262 that the thing is precise and always returns to exactly the center position
263 (if it has any).
264
265 BITS_TO_LONGS(), BIT_WORD(), BIT_MASK()
266 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
267
268 These three macros from bitops.h help some bitfield computations::
269
270 BITS_TO_LONGS(x) - returns the length of a bitfield array in longs for
271 x bits
272 BIT_WORD(x) - returns the index in the array in longs for bit x
273 BIT_MASK(x) - returns the index in a long for bit x
274
275 The id* and name fields
276 ~~~~~~~~~~~~~~~~~~~~~~~
277
278 The dev->name should be set before registering the input device by the input
279 device driver. It's a string like 'Generic button device' containing a
280 user friendly name of the device.
281
282 The id* fields contain the bus ID (PCI, USB, ...), vendor ID and device ID
283 of the device. The bus IDs are defined in input.h. The vendor and device IDs
284 are defined in pci_ids.h, usb_ids.h and similar include files. These fields
285 should be set by the input device driver before registering it.
286
287 The idtype field can be used for specific information for the input device
288 driver.
289
290 The id and name fields can be passed to userland via the evdev interface.
291
292 The keycode, keycodemax, keycodesize fields
293 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
294
295 These three fields should be used by input devices that have dense keymaps.
296 The keycode is an array used to map from scancodes to input system keycodes.
297 The keycode max should contain the size of the array and keycodesize the
298 size of each entry in it (in bytes).
299
300 Userspace can query and alter current scancode to keycode mappings using
301 EVIOCGKEYCODE and EVIOCSKEYCODE ioctls on corresponding evdev interface.
302 When a device has all 3 aforementioned fields filled in, the driver may
303 rely on kernel's default implementation of setting and querying keycode
304 mappings.
305
306 dev->getkeycode() and dev->setkeycode()
307 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
308
309 getkeycode() and setkeycode() callbacks allow drivers to override default
310 keycode/keycodesize/keycodemax mapping mechanism provided by input core
311 and implement sparse keycode maps.
312
313 Key autorepeat
314 ~~~~~~~~~~~~~~
315
316 ... is simple. It is handled by the input.c module. Hardware autorepeat is
317 not used, because it's not present in many devices and even where it is
318 present, it is broken sometimes (at keyboards: Toshiba notebooks). To enable
319 autorepeat for your device, just set EV_REP in dev->evbit. All will be
320 handled by the input system.
321
322 Other event types, handling output events
323 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
324
325 The other event types up to now are:
326
327 - EV_LED - used for the keyboard LEDs.
328 - EV_SND - used for keyboard beeps.
329
330 They are very similar to for example key events, but they go in the other
331 direction - from the system to the input device driver. If your input device
332 driver can handle these events, it has to set the respective bits in evbit,
333 *and* also the callback routine::
334
335 button_dev->event = button_event;
336
337 int button_event(struct input_dev *dev, unsigned int type,
338 unsigned int code, int value)
339 {
340 if (type == EV_SND && code == SND_BELL) {
341 outb(value, BUTTON_BELL);
342 return 0;
343 }
344 return -1;
345 }
346
347 This callback routine can be called from an interrupt or a BH (although that
348 isn't a rule), and thus must not sleep, and must not take too long to finish.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.