Documentation / admin-guide / pnp.rst


Based on kernel version 6.11. Page generated on 2024-09-24 08:21 EST.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288
=================================
Linux Plug and Play Documentation
=================================

:Author: Adam Belay <ambx1@neo.rr.com>
:Last updated: Oct. 16, 2002


Overview
--------

Plug and Play provides a means of detecting and setting resources for legacy or
otherwise unconfigurable devices.  The Linux Plug and Play Layer provides these 
services to compatible drivers.


The User Interface
------------------

The Linux Plug and Play user interface provides a means to activate PnP devices
for legacy and user level drivers that do not support Linux Plug and Play.  The 
user interface is integrated into sysfs.

In addition to the standard sysfs file the following are created in each
device's directory:
- id - displays a list of support EISA IDs
- options - displays possible resource configurations
- resources - displays currently allocated resources and allows resource changes

activating a device
^^^^^^^^^^^^^^^^^^^

::

	# echo "auto" > resources

this will invoke the automatic resource config system to activate the device

manually activating a device
^^^^^^^^^^^^^^^^^^^^^^^^^^^^

::

	# echo "manual <depnum> <mode>" > resources

	<depnum> - the configuration number
	<mode> - static or dynamic
		 static = for next boot
		 dynamic = now

disabling a device
^^^^^^^^^^^^^^^^^^

::

	# echo "disable" > resources


EXAMPLE:

Suppose you need to activate the floppy disk controller.

1. change to the proper directory, in my case it is
   /driver/bus/pnp/devices/00:0f::

	# cd /driver/bus/pnp/devices/00:0f
	# cat name
	PC standard floppy disk controller

2. check if the device is already active::

	# cat resources
	DISABLED

  - Notice the string "DISABLED".  This means the device is not active.

3. check the device's possible configurations (optional)::

	# cat options
	Dependent: 01 - Priority acceptable
	    port 0x3f0-0x3f0, align 0x7, size 0x6, 16-bit address decoding
	    port 0x3f7-0x3f7, align 0x0, size 0x1, 16-bit address decoding
	    irq 6
	    dma 2 8-bit compatible
	Dependent: 02 - Priority acceptable
	    port 0x370-0x370, align 0x7, size 0x6, 16-bit address decoding
	    port 0x377-0x377, align 0x0, size 0x1, 16-bit address decoding
	    irq 6
	    dma 2 8-bit compatible

4. now activate the device::

	# echo "auto" > resources

5. finally check if the device is active::

	# cat resources
	io 0x3f0-0x3f5
	io 0x3f7-0x3f7
	irq 6
	dma 2

also there are a series of kernel parameters::

	pnp_reserve_irq=irq1[,irq2] ....
	pnp_reserve_dma=dma1[,dma2] ....
	pnp_reserve_io=io1,size1[,io2,size2] ....
	pnp_reserve_mem=mem1,size1[,mem2,size2] ....



The Unified Plug and Play Layer
-------------------------------

All Plug and Play drivers, protocols, and services meet at a central location
called the Plug and Play Layer.  This layer is responsible for the exchange of 
information between PnP drivers and PnP protocols.  Thus it automatically 
forwards commands to the proper protocol.  This makes writing PnP drivers 
significantly easier.

The following functions are available from the Plug and Play Layer:

pnp_get_protocol
  increments the number of uses by one

pnp_put_protocol
  deincrements the number of uses by one

pnp_register_protocol
  use this to register a new PnP protocol

pnp_unregister_protocol
  use this function to remove a PnP protocol from the Plug and Play Layer

pnp_register_driver
  adds a PnP driver to the Plug and Play Layer

  this includes driver model integration
  returns zero for success or a negative error number for failure; count
  calls to the .add() method if you need to know how many devices bind to
  the driver

pnp_unregister_driver
  removes a PnP driver from the Plug and Play Layer



Plug and Play Protocols
-----------------------

This section contains information for PnP protocol developers.

The following Protocols are currently available in the computing world:

- PNPBIOS:
    used for system devices such as serial and parallel ports.
- ISAPNP:
    provides PnP support for the ISA bus
- ACPI:
    among its many uses, ACPI provides information about system level
    devices.

It is meant to replace the PNPBIOS.  It is not currently supported by Linux
Plug and Play but it is planned to be in the near future.


Requirements for a Linux PnP protocol:
1. the protocol must use EISA IDs
2. the protocol must inform the PnP Layer of a device's current configuration

- the ability to set resources is optional but preferred.

The following are PnP protocol related functions:

pnp_add_device
  use this function to add a PnP device to the PnP layer

  only call this function when all wanted values are set in the pnp_dev
  structure

pnp_init_device
  call this to initialize the PnP structure

pnp_remove_device
  call this to remove a device from the Plug and Play Layer.
  it will fail if the device is still in use.
  automatically will free mem used by the device and related structures

pnp_add_id
  adds an EISA ID to the list of supported IDs for the specified device

For more information consult the source of a protocol such as
/drivers/pnp/pnpbios/core.c.



Linux Plug and Play Drivers
---------------------------

This section contains information for Linux PnP driver developers.

The New Way
^^^^^^^^^^^

1. first make a list of supported EISA IDS

   ex::

	static const struct pnp_id pnp_dev_table[] = {
		/* Standard LPT Printer Port */
		{.id = "PNP0400", .driver_data = 0},
		/* ECP Printer Port */
		{.id = "PNP0401", .driver_data = 0},
		{.id = ""}
	};

   Please note that the character 'X' can be used as a wild card in the function
   portion (last four characters).

   ex::

	/* Unknown PnP modems */
	{	"PNPCXXX",		UNKNOWN_DEV	},

   Supported PnP card IDs can optionally be defined.
   ex::

	static const struct pnp_id pnp_card_table[] = {
		{	"ANYDEVS",		0	},
		{	"",			0	}
	};

2. Optionally define probe and remove functions.  It may make sense not to
   define these functions if the driver already has a reliable method of detecting
   the resources, such as the parport_pc driver.

   ex::

	static int
	serial_pnp_probe(struct pnp_dev * dev, const struct pnp_id *card_id, const
			struct pnp_id *dev_id)
	{
	. . .

   ex::

	static void serial_pnp_remove(struct pnp_dev * dev)
	{
	. . .

   consult /drivers/serial/8250_pnp.c for more information.

3. create a driver structure

   ex::

	static struct pnp_driver serial_pnp_driver = {
		.name		= "serial",
		.card_id_table	= pnp_card_table,
		.id_table	= pnp_dev_table,
		.probe		= serial_pnp_probe,
		.remove		= serial_pnp_remove,
	};

   * name and id_table cannot be NULL.

4. register the driver

   ex::

	static int __init serial8250_pnp_init(void)
	{
		return pnp_register_driver(&serial_pnp_driver);
	}

The Old Way
^^^^^^^^^^^

A series of compatibility functions have been created to make it easy to convert
ISAPNP drivers.  They should serve as a temporary solution only.

They are as follows::

	struct pnp_dev *pnp_find_dev(struct pnp_card *card,
				     unsigned short vendor,
				     unsigned short function,
				     struct pnp_dev *from)