1 How To Write Linux PCI Drivers
2
3 by Martin Mares <mj@suse.cz> on 07-Feb-2000
4
5 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
6 The world of PCI is vast and it's full of (mostly unpleasant) surprises.
7 Different PCI devices have different requirements and different bugs --
8 because of this, the PCI support layer in Linux kernel is not as trivial
9 as one would wish. This short pamphlet tries to help all potential driver
10 authors to find their way through the deep forests of PCI handling.
11
12
13 0. Structure of PCI drivers
14 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
15 There exist two kinds of PCI drivers: new-style ones (which leave most of
16 probing for devices to the PCI layer and support online insertion and removal
17 of devices [thus supporting PCI, hot-pluggable PCI and CardBus in single
18 driver]) and old-style ones which just do all the probing themselves. Unless
19 you have a very good reason to do so, please don't use the old way of probing
20 in any new code. After the driver finds the devices it wishes to operate
21 on (either the old or the new way), it needs to perform the following steps:
22
23 Enable the device
24 Access device configuration space
25 Discover resources (addresses and IRQ numbers) provided by the device
26 Allocate these resources
27 Communicate with the device
28
29 Most of these topics are covered by the following sections, for the rest
30 look at <linux/pci.h>, it's hopefully well commented.
31
32 If the PCI subsystem is not configured (CONFIG_PCI is not set), most of
33 the functions described below are defined as inline functions either completely
34 empty or just returning an appropriate error codes to avoid lots of ifdefs
35 in the drivers.
36
37
38 1. New-style drivers
39 ~~~~~~~~~~~~~~~~~~~~
40 The new-style drivers just call pci_register_driver during their initialization
41 with a pointer to a structure describing the driver (struct pci_driver) which
42 contains:
43
44 name Name of the driver
45 id_table Pointer to table of device ID's the driver is
46 interested in. Most drivers should export this
47 table using MODULE_DEVICE_TABLE(pci,...).
48 Set to NULL to call probe() function for every
49 PCI device known to the system.
50 probe Pointer to a probing function which gets called (during
51 execution of pci_register_driver for already existing
52 devices or later if a new device gets inserted) for all
53 PCI devices which match the ID table and are not handled
54 by the other drivers yet. This function gets passed a pointer
55 to the pci_dev structure representing the device and also
56 which entry in the ID table did the device match. It returns
57 zero when the driver has accepted the device or an error
58 code (negative number) otherwise. This function always gets
59 called from process context, so it can sleep.
60 remove Pointer to a function which gets called whenever a device
61 being handled by this driver is removed (either during
62 deregistration of the driver or when it's manually pulled
63 out of a hot-pluggable slot). This function can be called
64 from interrupt context.
65 suspend, Power management hooks -- called when the device goes to
66 resume sleep or is resumed.
67
68 The ID table is an array of struct pci_device_id ending with a all-zero entry.
69 Each entry consists of:
70
71 vendor, device Vendor and device ID to match (or PCI_ANY_ID)
72 subvendor, Subsystem vendor and device ID to match (or PCI_ANY_ID)
73 subdevice
74 class, Device class to match. The class_mask tells which bits
75 class_mask of the class are honored during the comparison.
76 driver_data Data private to the driver.
77
78 When the driver exits, it just calls pci_deregister_driver() and the PCI layer
79 automatically calls the remove hook for all devices handled by the driver.
80
81 Please mark the initialization and cleanup functions where appropriate
82 (the corresponding macros are defined in <linux/init.h>):
83
84 __init Initialization code. Thrown away after the driver
85 initializes.
86 __exit Exit code. Ignored for non-modular drivers.
87 __devinit Device initialization code. Identical to __init if
88 the kernel is not compiled with CONFIG_HOTPLUG, normal
89 function otherwise.
90 __devexit The same for __exit.
91
92 Tips:
93 The module_init()/module_exit() functions (and all initialization
94 functions called only from these) should be marked __init/exit.
95 The struct pci_driver shouldn't be marked with any of these tags.
96 The ID table array should be marked __devinitdata.
97 The probe() and remove() functions (and all initialization
98 functions called only from these) should be marked __devinit/exit.
99 If you are sure the driver is not a hotplug driver then use only
100 __init/exit __initdata/exitdata.
101
102
103 2. How to find PCI devices manually (the old style)
104 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
105 PCI drivers not using the pci_register_driver() interface search
106 for PCI devices manually using the following constructs:
107
108 Searching by vendor and device ID:
109
110 struct pci_dev *dev = NULL;
111 while (dev = pci_find_device(VENDOR_ID, DEVICE_ID, dev))
112 configure_device(dev);
113
114 Searching by class ID (iterate in a similar way):
115
116 pci_find_class(CLASS_ID, dev)
117
118 Searching by both vendor/device and subsystem vendor/device ID:
119
120 pci_find_subsys(VENDOR_ID, DEVICE_ID, SUBSYS_VENDOR_ID, SUBSYS_DEVICE_ID, dev).
121
122 You can use the constant PCI_ANY_ID as a wildcard replacement for
123 VENDOR_ID or DEVICE_ID. This allows searching for any device from a
124 specific vendor, for example.
125
126 In case you need to decide according to some more complex criteria,
127 you can walk the list of all known PCI devices yourself:
128
129 struct pci_dev *dev;
130 pci_for_each_dev(dev) {
131 ... do anything you want with dev ...
132 }
133
134 For compatibility with device ordering in older kernels, you can also
135 use pci_for_each_dev_reverse(dev) for walking the list in the opposite
136 direction.
137
138
139 3. Enabling devices
140 ~~~~~~~~~~~~~~~~~~~
141 Before you do anything with the device you've found, you need to enable
142 it by calling pci_enable_device() which enables I/O and memory regions of
143 the device, assigns missing resources if needed and wakes up the device
144 if it was in suspended state. Please note that this function can fail.
145
146 If you want to use the device in bus mastering mode, call pci_set_master()
147 which enables the bus master bit in PCI_COMMAND register and also fixes
148 the latency timer value if it's set to something bogus by the BIOS.
149
150
151 4. How to access PCI config space
152 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
153 You can use pci_(read|write)_config_(byte|word|dword) to access the config
154 space of a device represented by struct pci_dev *. All these functions return 0
155 when successful or an error code (PCIBIOS_...) which can be translated to a text
156 string by pcibios_strerror. Most drivers expect that accesses to valid PCI
157 devices don't fail.
158
159 If you access fields in the standard portion of the config header, please
160 use symbolic names of locations and bits declared in <linux/pci.h>.
161
162 If you need to access Extended PCI Capability registers, just call
163 pci_find_capability() for the particular capability and it will find the
164 corresponding register block for you.
165
166
167 5. Addresses and interrupts
168 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
169 Memory and port addresses and interrupt numbers should NOT be read from the
170 config space. You should use the values in the pci_dev structure as they might
171 have been remapped by the kernel.
172
173 See Documentation/IO-mapping.txt for how to access device memory.
174
175 You still need to call request_region() for I/O regions and request_mem_region()
176 for memory regions to make sure nobody else is using the same device.
177
178 All interrupt handlers should be registered with SA_SHIRQ and use the devid
179 to map IRQs to devices (remember that all PCI interrupts are shared).
180
181
182 6. Other interesting functions
183 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
184 pci_find_slot() Find pci_dev corresponding to given bus and
185 slot numbers.
186 pci_set_power_state() Set PCI Power Management state (0=D0 ... 3=D3)
187 pci_find_capability() Find specified capability in device's capability
188 list.
189 pci_module_init() Inline helper function for ensuring correct
190 pci_driver initialization and error handling.
191 pci_resource_start() Returns bus start address for a given PCI region
192 pci_resource_end() Returns bus end address for a given PCI region
193 pci_resource_len() Returns the byte length of a PCI region
194 pci_set_drvdata() Set private driver data pointer for a pci_dev
195 pci_get_drvdata() Return private driver data pointer for a pci_dev
196
197
198 7. Miscellaneous hints
199 ~~~~~~~~~~~~~~~~~~~~~~
200 When displaying PCI slot names to the user (for example when a driver wants
201 to tell the user what card has it found), please use pci_dev->slot_name
202 for this purpose.
203
204 Always refer to the PCI devices by a pointer to the pci_dev structure.
205 All PCI layer functions use this identification and it's the only
206 reasonable one. Don't use bus/slot/function numbers except for very
207 special purposes -- on systems with multiple primary buses their semantics
208 can be pretty complex.
209
210 If you're going to use PCI bus mastering DMA, take a look at
211 Documentation/DMA-mapping.txt.
212
213
214 8. Obsolete functions
215 ~~~~~~~~~~~~~~~~~~~~~
216 There are several functions kept only for compatibility with old drivers
217 not updated to the new PCI interface. Please don't use them in new code.
218
219 pcibios_present() Since ages, you don't need to test presence
220 of PCI subsystem when trying to talk with it.
221 If it's not there, the list of PCI devices
222 is empty and all functions for searching for
223 devices just return NULL.
224 pcibios_(read|write)_* Superseded by their pci_(read|write)_*
225 counterparts.
226 pcibios_find_* Superseded by their pci_find_* counterparts.
This page was automatically generated by the
LXR engine.
Visit the LXR main site for more
information.