1 ============== 1 ==============
2 Device Drivers 2 Device Drivers
3 ============== 3 ==============
4 4
5 See the kerneldoc for the struct device_driver 5 See the kerneldoc for the struct device_driver.
6 6
>> 7
7 Allocation 8 Allocation
8 ~~~~~~~~~~ 9 ~~~~~~~~~~
9 10
10 Device drivers are statically allocated struct 11 Device drivers are statically allocated structures. Though there may
11 be multiple devices in a system that a driver 12 be multiple devices in a system that a driver supports, struct
12 device_driver represents the driver as a whole 13 device_driver represents the driver as a whole (not a particular
13 device instance). 14 device instance).
14 15
15 Initialization 16 Initialization
16 ~~~~~~~~~~~~~~ 17 ~~~~~~~~~~~~~~
17 18
18 The driver must initialize at least the name a 19 The driver must initialize at least the name and bus fields. It should
19 also initialize the devclass field (when it ar 20 also initialize the devclass field (when it arrives), so it may obtain
20 the proper linkage internally. It should also 21 the proper linkage internally. It should also initialize as many of
21 the callbacks as possible, though each is opti 22 the callbacks as possible, though each is optional.
22 23
23 Declaration 24 Declaration
24 ~~~~~~~~~~~ 25 ~~~~~~~~~~~
25 26
26 As stated above, struct device_driver objects 27 As stated above, struct device_driver objects are statically
27 allocated. Below is an example declaration of 28 allocated. Below is an example declaration of the eepro100
28 driver. This declaration is hypothetical only; 29 driver. This declaration is hypothetical only; it relies on the driver
29 being converted completely to the new model:: 30 being converted completely to the new model::
30 31
31 static struct device_driver eepro100_driver 32 static struct device_driver eepro100_driver = {
32 .name = "eepro100", 33 .name = "eepro100",
33 .bus = &pci_bus_type, 34 .bus = &pci_bus_type,
34 35
35 .probe = eepro100_probe, 36 .probe = eepro100_probe,
36 .remove = eepro100_rem 37 .remove = eepro100_remove,
37 .suspend = eepro100_sus 38 .suspend = eepro100_suspend,
38 .resume = eepro100_res 39 .resume = eepro100_resume,
39 }; 40 };
40 41
41 Most drivers will not be able to be converted 42 Most drivers will not be able to be converted completely to the new
42 model because the bus they belong to has a bus 43 model because the bus they belong to has a bus-specific structure with
43 bus-specific fields that cannot be generalized 44 bus-specific fields that cannot be generalized.
44 45
45 The most common example of this are device ID 46 The most common example of this are device ID structures. A driver
46 typically defines an array of device IDs that 47 typically defines an array of device IDs that it supports. The format
47 of these structures and the semantics for comp 48 of these structures and the semantics for comparing device IDs are
48 completely bus-specific. Defining them as bus- 49 completely bus-specific. Defining them as bus-specific entities would
49 sacrifice type-safety, so we keep bus-specific 50 sacrifice type-safety, so we keep bus-specific structures around.
50 51
51 Bus-specific drivers should include a generic 52 Bus-specific drivers should include a generic struct device_driver in
52 the definition of the bus-specific driver. Lik 53 the definition of the bus-specific driver. Like this::
53 54
54 struct pci_driver { 55 struct pci_driver {
55 const struct pci_device_id *id_table; 56 const struct pci_device_id *id_table;
56 struct device_driver driver; 57 struct device_driver driver;
57 }; 58 };
58 59
59 A definition that included bus-specific fields 60 A definition that included bus-specific fields would look like
60 (using the eepro100 driver again):: 61 (using the eepro100 driver again)::
61 62
62 static struct pci_driver eepro100_driver = { 63 static struct pci_driver eepro100_driver = {
63 .id_table = eepro100_pci_tbl, 64 .id_table = eepro100_pci_tbl,
64 .driver = { 65 .driver = {
65 .name = "eepro100", 66 .name = "eepro100",
66 .bus = &pci_bus_typ 67 .bus = &pci_bus_type,
67 .probe = eepro100_pro 68 .probe = eepro100_probe,
68 .remove = eepro100_rem 69 .remove = eepro100_remove,
69 .suspend = eepro100_sus 70 .suspend = eepro100_suspend,
70 .resume = eepro100_res 71 .resume = eepro100_resume,
71 }, 72 },
72 }; 73 };
73 74
74 Some may find the syntax of embedded struct in 75 Some may find the syntax of embedded struct initialization awkward or
75 even a bit ugly. So far, it's the best way we' 76 even a bit ugly. So far, it's the best way we've found to do what we want...
76 77
77 Registration 78 Registration
78 ~~~~~~~~~~~~ 79 ~~~~~~~~~~~~
79 80
80 :: 81 ::
81 82
82 int driver_register(struct device_driver *dr 83 int driver_register(struct device_driver *drv);
83 84
84 The driver registers the structure on startup. 85 The driver registers the structure on startup. For drivers that have
85 no bus-specific fields (i.e. don't have a bus- 86 no bus-specific fields (i.e. don't have a bus-specific driver
86 structure), they would use driver_register and 87 structure), they would use driver_register and pass a pointer to their
87 struct device_driver object. 88 struct device_driver object.
88 89
89 Most drivers, however, will have a bus-specifi 90 Most drivers, however, will have a bus-specific structure and will
90 need to register with the bus using something 91 need to register with the bus using something like pci_driver_register.
91 92
92 It is important that drivers register their dr 93 It is important that drivers register their driver structure as early as
93 possible. Registration with the core initializ 94 possible. Registration with the core initializes several fields in the
94 struct device_driver object, including the ref 95 struct device_driver object, including the reference count and the
95 lock. These fields are assumed to be valid at 96 lock. These fields are assumed to be valid at all times and may be
96 used by the device model core or the bus drive 97 used by the device model core or the bus driver.
97 98
98 99
99 Transition Bus Drivers 100 Transition Bus Drivers
100 ~~~~~~~~~~~~~~~~~~~~~~ 101 ~~~~~~~~~~~~~~~~~~~~~~
101 102
102 By defining wrapper functions, the transition 103 By defining wrapper functions, the transition to the new model can be
103 made easier. Drivers can ignore the generic st 104 made easier. Drivers can ignore the generic structure altogether and
104 let the bus wrapper fill in the fields. For th 105 let the bus wrapper fill in the fields. For the callbacks, the bus can
105 define generic callbacks that forward the call 106 define generic callbacks that forward the call to the bus-specific
106 callbacks of the drivers. 107 callbacks of the drivers.
107 108
108 This solution is intended to be only temporary 109 This solution is intended to be only temporary. In order to get class
109 information in the driver, the drivers must be 110 information in the driver, the drivers must be modified anyway. Since
110 converting drivers to the new model should red 111 converting drivers to the new model should reduce some infrastructural
111 complexity and code size, it is recommended th 112 complexity and code size, it is recommended that they are converted as
112 class information is added. 113 class information is added.
113 114
114 Access 115 Access
115 ~~~~~~ 116 ~~~~~~
116 117
117 Once the object has been registered, it may ac 118 Once the object has been registered, it may access the common fields of
118 the object, like the lock and the list of devi 119 the object, like the lock and the list of devices::
119 120
120 int driver_for_each_dev(struct device_driver 121 int driver_for_each_dev(struct device_driver *drv, void *data,
121 int (*callback)(stru 122 int (*callback)(struct device *dev, void *data));
122 123
123 The devices field is a list of all the devices 124 The devices field is a list of all the devices that have been bound to
124 the driver. The LDM core provides a helper fun 125 the driver. The LDM core provides a helper function to operate on all
125 the devices a driver controls. This helper loc 126 the devices a driver controls. This helper locks the driver on each
126 node access, and does proper reference countin 127 node access, and does proper reference counting on each device as it
127 accesses it. 128 accesses it.
128 129
129 130
130 sysfs 131 sysfs
131 ~~~~~ 132 ~~~~~
132 133
133 When a driver is registered, a sysfs directory 134 When a driver is registered, a sysfs directory is created in its
134 bus's directory. In this directory, the driver 135 bus's directory. In this directory, the driver can export an interface
135 to userspace to control operation of the drive 136 to userspace to control operation of the driver on a global basis;
136 e.g. toggling debugging output in the driver. 137 e.g. toggling debugging output in the driver.
137 138
138 A future feature of this directory will be a ' 139 A future feature of this directory will be a 'devices' directory. This
139 directory will contain symlinks to the directo 140 directory will contain symlinks to the directories of devices it
140 supports. 141 supports.
141 142
142 143
143 144
144 Callbacks 145 Callbacks
145 ~~~~~~~~~ 146 ~~~~~~~~~
146 147
147 :: 148 ::
148 149
149 int (*probe) (struct device 150 int (*probe) (struct device *dev);
150 151
151 The probe() entry is called in task context, w 152 The probe() entry is called in task context, with the bus's rwsem locked
152 and the driver partially bound to the device. 153 and the driver partially bound to the device. Drivers commonly use
153 container_of() to convert "dev" to a bus-speci 154 container_of() to convert "dev" to a bus-specific type, both in probe()
154 and other routines. That type often provides 155 and other routines. That type often provides device resource data, such
155 as pci_dev.resource[] or platform_device.resou 156 as pci_dev.resource[] or platform_device.resources, which is used in
156 addition to dev->platform_data to initialize t 157 addition to dev->platform_data to initialize the driver.
157 158
158 This callback holds the driver-specific logic 159 This callback holds the driver-specific logic to bind the driver to a
159 given device. That includes verifying that th 160 given device. That includes verifying that the device is present, that
160 it's a version the driver can handle, that dri 161 it's a version the driver can handle, that driver data structures can
161 be allocated and initialized, and that any har 162 be allocated and initialized, and that any hardware can be initialized.
162 Drivers often store a pointer to their state w 163 Drivers often store a pointer to their state with dev_set_drvdata().
163 When the driver has successfully bound itself 164 When the driver has successfully bound itself to that device, then probe()
164 returns zero and the driver model code will fi 165 returns zero and the driver model code will finish its part of binding
165 the driver to that device. 166 the driver to that device.
166 167
167 A driver's probe() may return a negative errno 168 A driver's probe() may return a negative errno value to indicate that
168 the driver did not bind to this device, in whi 169 the driver did not bind to this device, in which case it should have
169 released all resources it allocated. !! 170 released all resources it allocated::
170 <<
171 Optionally, probe() may return -EPROBE_DEFER i <<
172 resources that are not yet available (e.g., su <<
173 hasn't initialized yet). The driver core will <<
174 deferred probe list and will try to call it ag <<
175 must defer, it should return -EPROBE_DEFER as <<
176 reduce the amount of time spent on setup work <<
177 unwound and reexecuted at a later time. <<
178 <<
179 .. warning:: <<
180 -EPROBE_DEFER must not be returned if pr <<
181 child devices, even if those child devic <<
182 in a cleanup path. If -EPROBE_DEFER is r <<
183 device has been registered, it may resul <<
184 .probe() calls to the same driver. <<
185 <<
186 :: <<
187 171
188 void (*sync_state) (struct device !! 172 void (*sync_state)(struct device *dev);
189 173
190 sync_state is called only once for a device. I 174 sync_state is called only once for a device. It's called when all the consumer
191 devices of the device have successfully probed 175 devices of the device have successfully probed. The list of consumers of the
192 device is obtained by looking at the device li 176 device is obtained by looking at the device links connecting that device to its
193 consumer devices. 177 consumer devices.
194 178
195 The first attempt to call sync_state() is made 179 The first attempt to call sync_state() is made during late_initcall_sync() to
196 give firmware and drivers time to link devices 180 give firmware and drivers time to link devices to each other. During the first
197 attempt at calling sync_state(), if all the co 181 attempt at calling sync_state(), if all the consumers of the device at that
198 point in time have already probed successfully 182 point in time have already probed successfully, sync_state() is called right
199 away. If there are no consumers of the device 183 away. If there are no consumers of the device during the first attempt, that
200 too is considered as "all consumers of the dev 184 too is considered as "all consumers of the device have probed" and sync_state()
201 is called right away. 185 is called right away.
202 186
203 If during the first attempt at calling sync_st 187 If during the first attempt at calling sync_state() for a device, there are
204 still consumers that haven't probed successful 188 still consumers that haven't probed successfully, the sync_state() call is
205 postponed and reattempted in the future only w 189 postponed and reattempted in the future only when one or more consumers of the
206 device probe successfully. If during the reatt 190 device probe successfully. If during the reattempt, the driver core finds that
207 there are one or more consumers of the device 191 there are one or more consumers of the device that haven't probed yet, then
208 sync_state() call is postponed again. 192 sync_state() call is postponed again.
209 193
210 A typical use case for sync_state() is to have 194 A typical use case for sync_state() is to have the kernel cleanly take over
211 management of devices from the bootloader. For 195 management of devices from the bootloader. For example, if a device is left on
212 and at a particular hardware configuration by 196 and at a particular hardware configuration by the bootloader, the device's
213 driver might need to keep the device in the bo 197 driver might need to keep the device in the boot configuration until all the
214 consumers of the device have probed. Once all 198 consumers of the device have probed. Once all the consumers of the device have
215 probed, the device's driver can synchronize th 199 probed, the device's driver can synchronize the hardware state of the device to
216 match the aggregated software state requested 200 match the aggregated software state requested by all the consumers. Hence the
217 name sync_state(). 201 name sync_state().
218 202
219 While obvious examples of resources that can b 203 While obvious examples of resources that can benefit from sync_state() include
220 resources such as regulator, sync_state() can 204 resources such as regulator, sync_state() can also be useful for complex
221 resources like IOMMUs. For example, IOMMUs wit 205 resources like IOMMUs. For example, IOMMUs with multiple consumers (devices
222 whose addresses are remapped by the IOMMU) mig 206 whose addresses are remapped by the IOMMU) might need to keep their mappings
223 fixed at (or additive to) the boot configurati 207 fixed at (or additive to) the boot configuration until all its consumers have
224 probed. 208 probed.
225 209
226 While the typical use case for sync_state() is 210 While the typical use case for sync_state() is to have the kernel cleanly take
227 over management of devices from the bootloader 211 over management of devices from the bootloader, the usage of sync_state() is
228 not restricted to that. Use it whenever it mak 212 not restricted to that. Use it whenever it makes sense to take an action after
229 all the consumers of a device have probed:: 213 all the consumers of a device have probed::
230 214
231 int (*remove) (struct device 215 int (*remove) (struct device *dev);
232 216
233 remove is called to unbind a driver from a dev 217 remove is called to unbind a driver from a device. This may be
234 called if a device is physically removed from 218 called if a device is physically removed from the system, if the
235 driver module is being unloaded, during a rebo 219 driver module is being unloaded, during a reboot sequence, or
236 in other cases. 220 in other cases.
237 221
238 It is up to the driver to determine if the dev 222 It is up to the driver to determine if the device is present or
239 not. It should free any resources allocated sp 223 not. It should free any resources allocated specifically for the
240 device; i.e. anything in the device's driver_d 224 device; i.e. anything in the device's driver_data field.
241 225
242 If the device is still present, it should quie 226 If the device is still present, it should quiesce the device and place
243 it into a supported low-power state. !! 227 it into a supported low-power state::
244 <<
245 :: <<
246 228
247 int (*suspend) (struct device 229 int (*suspend) (struct device *dev, pm_message_t state);
248 230
249 suspend is called to put the device in a low p !! 231 suspend is called to put the device in a low power state::
250 <<
251 :: <<
252 232
253 int (*resume) (struct device 233 int (*resume) (struct device *dev);
254 234
255 Resume is used to bring a device back from a l 235 Resume is used to bring a device back from a low power state.
256 236
257 237
258 Attributes 238 Attributes
259 ~~~~~~~~~~ 239 ~~~~~~~~~~
260 240
261 :: 241 ::
262 242
263 struct driver_attribute { 243 struct driver_attribute {
264 struct attribute attr; 244 struct attribute attr;
265 ssize_t (*show)(struct device_driver 245 ssize_t (*show)(struct device_driver *driver, char *buf);
266 ssize_t (*store)(struct device_drive 246 ssize_t (*store)(struct device_driver *, const char *buf, size_t count);
267 }; 247 };
268 248
269 Device drivers can export attributes via their 249 Device drivers can export attributes via their sysfs directories.
270 Drivers can declare attributes using a DRIVER_ 250 Drivers can declare attributes using a DRIVER_ATTR_RW and DRIVER_ATTR_RO
271 macro that works identically to the DEVICE_ATT 251 macro that works identically to the DEVICE_ATTR_RW and DEVICE_ATTR_RO
272 macros. 252 macros.
273 253
274 Example:: 254 Example::
275 255
276 DRIVER_ATTR_RW(debug); 256 DRIVER_ATTR_RW(debug);
277 257
278 This is equivalent to declaring:: 258 This is equivalent to declaring::
279 259
280 struct driver_attribute driver_attr_de 260 struct driver_attribute driver_attr_debug;
281 261
282 This can then be used to add and remove the at 262 This can then be used to add and remove the attribute from the
283 driver's directory using:: 263 driver's directory using::
284 264
285 int driver_create_file(struct device_driver 265 int driver_create_file(struct device_driver *, const struct driver_attribute *);
286 void driver_remove_file(struct device_driver 266 void driver_remove_file(struct device_driver *, const struct driver_attribute *);
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.