1lspci(8) The PCI Utilities lspci(8)
2
6 lspci - list all PCI devices
7
9 lspci [options]
10
12 lspci is a utility for displaying information about PCI buses in the
13 system and devices connected to them.
14 By default, it shows a brief list of devices. Use the options described
16 below to request either a more verbose output or output intended for
17 parsing by other programs.
18 If you are going to report bugs in PCI device drivers or in lspci it‐
20 self, please include output of "lspci -vvx" or even better "lspci
21 -vvxxx" (however, see below for possible caveats).
22 Some parts of the output, especially in the highly verbose modes, are
24 probably intelligible only to experienced PCI hackers. For exact defi‐
25 nitions of the fields, please consult either the PCI specifications or
26 the header.h and /usr/include/linux/pci.h include files.
27 Access to some parts of the PCI configuration space is restricted to
29 root on many operating systems, so the features of lspci available to
30 normal users are limited. However, lspci tries its best to display as
31 much as available and mark all other information with <access denied>
32 text.
33
36 Basic display modes
37 -m Dump PCI device data in a backward-compatible machine readable
38 form. See below for details.
39 -mm Dump PCI device data in a machine readable form for easy parsing
41 by scripts. See below for details.
42 -t Show a tree-like diagram containing all buses, bridges, devices
44 and connections between them.
45 Display options
48 -v Be verbose and display detailed information about all devices.
49 -vv Be very verbose and display more details. This level includes
51 everything deemed useful.
52 -vvv Be even more verbose and display everything we are able to
54 parse, even if it doesn't look interesting at all (e.g., unde‐
55 fined memory regions).
56 -k Show kernel drivers handling each device and also kernel modules
58 capable of handling it. Turned on by default when -v is given
59 in the normal mode of output. (Currently works only on Linux
60 with kernel 2.6 or newer.)
61 -x Show hexadecimal dump of the standard part of the configuration
63 space (the first 64 bytes or 128 bytes for CardBus bridges).
64 -xxx Show hexadecimal dump of the whole PCI configuration space. It
66 is available only to root as several PCI devices crash when you
67 try to read some parts of the config space (this behavior proba‐
68 bly doesn't violate the PCI standard, but it's at least very
69 stupid). However, such devices are rare, so you needn't worry
70 much.
71 -xxxx Show hexadecimal dump of the extended (4096-byte) PCI configura‐
73 tion space available on PCI-X 2.0 and PCI Express buses.
74 -b Bus-centric view. Show all IRQ numbers and addresses as seen by
76 the cards on the PCI bus instead of as seen by the kernel.
77 -D Always show PCI domain numbers. By default, lspci suppresses
79 them on machines which have only domain 0.
80 -P Identify PCI devices by path through each bridge, instead of by
82 bus number.
83 -PP Identify PCI devices by path through each bridge, showing the
85 bus number as well as the device number.
86 Options to control resolving ID's to names
89 -n Show PCI vendor and device codes as numbers instead of looking
90 them up in the PCI ID list.
91 -nn Show PCI vendor and device codes as both numbers and names.
93 -q Use DNS to query the central PCI ID database if a device is not
95 found in the local pci.ids file. If the DNS query succeeds, the
96 result is cached in ~/.pciids-cache and it is recognized in sub‐
97 sequent runs even if -q is not given any more. Please use this
98 switch inside automated scripts only with caution to avoid over‐
99 loading the database servers.
100 -qq Same as -q, but the local cache is reset.
102 -Q Query the central database even for entries which are recognized
104 locally. Use this if you suspect that the displayed entry is
105 wrong.
106 Options for selection of devices
109 -s [[[[<domain>]:]<bus>]:][<device>][.[<func>]]
110 Show only devices in the specified domain (in case your machine
111 has several host bridges, they can either share a common bus
112 number space or each of them can address a PCI domain of its
113 own; domains are numbered from 0 to ffff), bus (0 to ff), device
114 (0 to 1f) and function (0 to 7). Each component of the device
115 address can be omitted or set to "*", both meaning "any value".
116 All numbers are hexadecimal. E.g., "0:" means all devices on
117 bus 0, "0" means all functions of device 0 on any bus, "0.3" se‐
118 lects third function of device 0 on all buses and ".4" shows
119 only the fourth function of each device.
120 -d [<vendor>]:[<device>][:<class>[:<prog-if>]]
122 Show only devices with specified vendor, device, class ID, and
123 programming interface. The ID's are given in hexadecimal and
124 may be omitted or given as "*", both meaning "any value". The
125 class ID can contain "x" characters which stand for "any digit".
126 Other options
129 -i <file>
130 Use <file> as the PCI ID list instead of /usr/share/hw‐
131 data/pci.ids.
132 -p <file>
134 Use <file> as the map of PCI ID's handled by kernel modules. By
135 default, lspci uses /lib/modules/kernel_version/modules.pcimap.
136 Applies only to Linux systems with recent enough module tools.
137 -M Invoke bus mapping mode which performs a thorough scan of all
139 PCI devices, including those behind misconfigured bridges, etc.
140 This option gives meaningful results only with a direct hardware
141 access mode, which usually requires root privileges. By de‐
142 fault, the bus mapper scans domain. You can use the -s option to
143 select a different domain.
144 --version
146 Shows lspci version. This option should be used stand-alone.
147 PCI access options
150 The PCI utilities use the PCI library to talk to PCI devices (see
151 pcilib(7) for details). You can use the following options to influence
152 its behavior:
153 -A <method>
155 The library supports a variety of methods to access the PCI
156 hardware. By default, it uses the first access method avail‐
157 able, but you can use this option to override this decision. See
158 -A help for a list of available methods and their descriptions.
159 -O <param>=<value>
161 The behavior of the library is controlled by several named pa‐
162 rameters. This option allows one to set the value of any of the
163 parameters. Use -O help for a list of known parameters and their
164 default values.
165 -H1 Use direct hardware access via Intel configuration mechanism 1.
167 (This is a shorthand for -A intel-conf1.)
168 -H2 Use direct hardware access via Intel configuration mechanism 2.
170 (This is a shorthand for -A intel-conf2.)
171 -F <file>
173 Instead of accessing real hardware, read the list of devices and
174 values of their configuration registers from the given file pro‐
175 duced by an earlier run of lspci -x. This is very useful for
176 analysis of user-supplied bug reports, because you can display
177 the hardware configuration in any way you want without disturb‐
178 ing the user with requests for more dumps.
179 -G Increase debug level of the library.
181
184 If you intend to process the output of lspci automatically, please use
185 one of the machine-readable output formats (-m, -vm, -vmm) described in
186 this section. All other formats are likely to change between versions
187 of lspci.
188 All numbers are always printed in hexadecimal. If you want to process
191 numeric ID's instead of names, please add the -n switch.
192 Simple format (-m)
195 In the simple format, each device is described on a single line, which
196 is formatted as parameters suitable for passing to a shell script,
197 i.e., values separated by whitespaces, quoted and escaped if necessary.
198 Some of the arguments are positional: slot, class, vendor name, device
199 name, subsystem vendor name and subsystem name (the last two are empty
200 if the device has no subsystem); the remaining arguments are option-
201 like:
202 -rrev Revision number.
205 -pprogif
208 Programming interface.
209 The relative order of positional arguments and options is undefined.
212 New options can be added in future versions, but they will always have
213 a single argument not separated from the option by any spaces, so they
214 can be easily ignored if not recognized.
215 Verbose format (-vmm)
218 The verbose output is a sequence of records separated by blank lines.
219 Each record describes a single device by a sequence of lines, each line
220 containing a single `tag: value' pair. The tag and the value are sepa‐
221 rated by a single tab character. Neither the records nor the lines
222 within a record are in any particular order. Tags are case-sensitive.
223 The following tags are defined:
226 Slot The name of the slot where the device resides ([domain:]bus:de‐
229 vice.function). This tag is always the first in a record.
230 Class Name of the class.
233 Vendor Name of the vendor.
236 Device Name of the device.
239 SVendor
242 Name of the subsystem vendor (optional).
243 SDevice
246 Name of the subsystem (optional).
247 PhySlot
250 The physical slot where the device resides (optional, Linux
251 only).
252 Rev Revision number (optional).
255 ProgIf Programming interface (optional).
258 Driver Kernel driver currently handling the device (optional, Linux
261 only).
262 Module Kernel module reporting that it is capable of handling the de‐
265 vice (optional, Linux only). Multiple lines with this tag can
266 occur.
267 NUMANode
270 NUMA node this device is connected to (optional, Linux only).
271 IOMMUGroup
274 IOMMU group that this device is part of (optional, Linux only).
275 New tags can be added in future versions, so you should silently ignore
278 any tags you don't recognize.
279 Backward-compatible verbose format (-vm)
282 In this mode, lspci tries to be perfectly compatible with its old ver‐
283 sions. It's almost the same as the regular verbose format, but the De‐
284 vice tag is used for both the slot and the device name, so it occurs
285 twice in a single record. Please avoid using this format in any new
286 code.
287
290 /usr/share/hwdata/pci.ids
291 A list of all known PCI ID's (vendors, devices, classes and sub‐
292 classes). Maintained at https://pci-ids.ucw.cz/, use the update-
293 pciids utility to download the most recent version.
294 /usr/share/hwdata/pci.ids.gz
296 If lspci is compiled with support for compression, this file is
297 tried before pci.ids.
298 ~/.pciids-cache
300 All ID's found in the DNS query mode are cached in this file.
301
304 Sometimes, lspci is not able to decode the configuration registers com‐
305 pletely. This usually happens when not enough documentation was avail‐
306 able to the authors. In such cases, it at least prints the <?> mark to
307 signal that there is potentially something more to say. If you know the
308 details, patches will be of course welcome.
309 Access to the extended configuration space is currently supported only
311 by the linux_sysfs back-end.
312
315 setpci(8), pci.ids(5), update-pciids(8), pcilib(7)
316
319 The PCI Utilities are maintained by Martin Mares <mj@ucw.cz>.
320pciutils-3.10.0 01 May 2023 lspci(8)