> Linux Man Page: "lspci"


Search with Google


The Linux Documentation Project maintains an archive of snaphots of the (English language) core Linux manual pages that are maintained by Michael Kerrisk. Corrections and additions are welcome, but review the "Help Wanted" list, first.

Man pages belonging to programs are usually distributed together with those programs. Therefore, the core Linux man-pages mainly contains the pages for system calls and library routines, special devices, and file formats. However, it also contains documentation for a few programs, in cases where the authors or maintainers of the program do not distribute man pages themselves.

This page is part of release 3.11 of the Linux man-pages project. A description of the project, and information about reporting bugs, can be found at


man page(1) manual page Table of Contents


lspci - list all PCI devices


lspci [options]


lspci is a utility for displaying information about all PCI buses in the system and all devices connected to them.

By default, it shows a brief list of devices. Use the options described below to request either a more verbose output or output intended for parsing by other programs.

If you are going to report bugs in PCI device drivers or in lspci itself, please include output of “lspci -vvx” or even better “lspci -vvxxx” (however, see below for possible caveats).

Some parts of the output, especially in the highly verbose modes, is probably intelligible only to experienced PCI hackers. For the exact definitions of the fields, please consult either the PCI specifications or the header.h and /usr/include/linux/pci.h include files.

Access to some parts of the PCI configuration space is restricted to root on many operating systems, so the features of lspci available to normal users are limited. However, lspci tries its best to display as much as available and mark all other information with <access denied> text.


Be verbose and display detailed information about all devices.

Be very verbose and display more details. This level includes everything deemed useful.

Be even more verbose and display everything we are able to parse, even if it doesn’t look interesting at all (e.g., undefined memory regions).

Show PCI vendor and device codes as numbers instead of looking them up in the PCI ID list.

Show PCI vendor and device codes as both numbers and names.

Show hexadecimal dump of the standard part of the configuration space (the first 64 bytes or 128 bytes for CardBus bridges).

Show hexadecimal dump of the whole PCI configuration space. It is available only to root as several PCI devices crash when you try to read some parts of the config space (this behavior probably doesn’t violate the PCI standard, but it’s at least very stupid). However, such devices are rare, so you needn’t worry much.

Show hexadecimal dump of the extended (4096-byte) PCI configuration space available on PCI-X 2.0 and PCI Express buses.

Show kernel drivers handling each device and also kernel modules capable of handling it. Turned on by default when -v is given in the normal mode of output. (Currently works only on Linux with kernel 2.6 or newer.)

Bus-centric view. Show all IRQ numbers and addresses as seen by the cards on the PCI bus instead of as seen by the kernel.

Show a tree-like diagram containing all buses, bridges, devices and connections between them.

-s [[[[<domain>]:]<bus>]:][<slot>][.[<func>]]
Show only devices in the specified domain (in case your machine has several host bridges, they can either share a common bus number space or each of them can address a PCI domain of its own; domains are numbered from 0 to ffff), bus (0 to ff), slot (0 to 1f) and function (0 to 7). Each component of the device address can be omitted or set to “*", both meaning “any value". All numbers are hexadecimal. E.g., “0:” means all devices on bus 0, “0” means all functions of device 0 on any bus, “0.3" selects third function of device 0 on all buses and “.4” shows only the fourth function of each device.

-d [<vendor>]:[<device>]
Show only devices with specified vendor and device ID. Both ID’s are given in hexadecimal and may be omitted or given as “*", both meaning “any value".

-i <file>
Use <file> as the PCI ID list instead of /usr/share/hwdata/pci.ids.

-p <file>
Use <file> as the map of PCI ID’s handled by kernel modules. By default, lspci uses /lib/modules/kernel_version/modules.pcimap. Applies only to Linux systems with recent enough module tools.

Dump PCI device data in a backward-compatible machine readable form. See below for details.

Dump PCI device data in a machine readable form for easy parsing by scripts. See below for details.

Always show PCI domain numbers. By default, lspci suppresses them on machines which have only domain 0.

Invoke bus mapping mode which performs a thorough scan of all PCI devices, including those behind misconfigured bridges etc. This option is available only to root and it gives meaningful results only if combined with direct hardware access mode (otherwise the results are identical to normal listing modes, modulo bugs in lspci). Please note that the bus mapper doesn’t support PCI domains and scans only domain 0.

Shows lspci version. This option should be used stand-alone.

Pcilib and Its Options

The PCI utilities use PCILIB (a portable library providing platformindependent functions for PCI configuration space access) to talk to the PCI cards. It supports the following access methods:

The /sys filesystem on Linux 2.6 and newer. The standard header of the config space is available to all users, the rest only to root. Supports extended configuration space, PCI domains and information on attached kernel drivers.

The /proc/bus/pci interface supported by Linux 2.1 and newer. The standard header of the config space is available to all users, the rest only to root.

Direct hardware access via Intel configuration mechanism 1. Available on i386 and compatibles on Linux, Solaris/x86, GNU Hurd and Windows. Requires root privileges.

Direct hardware access via Intel configuration mechanism 2. Available on i386 and compatibles on Linux, Solaris/x86 and GNU Hurd. Requires root privileges. Warning: This method is able to address only first 16 devices on any bus and it seems to be very unreliable in many cases.

The /dev/pci device on FreeBSD. Requires root privileges.

The /dev/pci device on OpenBSD. Requires root privileges.

The /dev/pci0 device on NetBSD accessed using the local libpci library.

Access method used on AIX. Requires root privileges.

By default, PCILIB uses the first available access method and displays no debugging messages, but you can use the following switches to control its behavior:

-P <dir>
Force use of the linux_proc access method, using <dir> instead of /proc/bus/pci.

Use direct hardware access via Intel configuration mechanism 1.

Use direct hardware access via Intel configuration mechanism 2.

-F <file>
Extract all information from given file containing output of lspci -x. This is very useful for analysis of user-supplied bug reports, because you can display the hardware configuration in any way you want without disturbing the user with requests for more dumps.

Increase debug level of the library.

Machine Readable Output

If you intend to process the output of lspci automatically, please use one of the machine-readable output formats (-m, -vm, -vmm) described in this section. All other formats are likely to change between versions of lspci.

All numbers are always printed in hexadecimal. If you want to process numeric ID’s instead of names, please add the -n switch.

Simple format (-m)
In the simple format, each device is described on a single line, which is formatted as parameters suitable for passing to a shell script, i.e., values separated by whitespaces, quoted and escaped if necessary. Some of the arguments are positional: slot, class, vendor name, device name, subsystem vendor name and subsystem name (the last two are empty if the device has no subsystem); the remaining arguments are optionlike:

Revision number.

Programming interface.

The relative order of positional arguments and options is undefined. New options can be added in future versions, but they will always have a single argument not separated from the option by any spaces, so they can be easily ignored if not recognized.

Verbose format (-vmm)
The verbose output is a sequence of records separated by blank lines. Each record describes a single device by a sequence of lines, each line containing a single ‘tag: value’ pair. The tag and the value are separated by a single tab character. Neither the records nor the lines within a record are in any particular order. Tags are case-sensitive.

The following tags are defined:

Slot The name of the slot where the device resides ([domain:]bus:device.function). This tag is always the first in a record.

Class Name of the class.

Vendor Name of the vendor.

Device Name of the device.

Name of the subsystem vendor (optional).

Name of the subsystem (optional).

Revision number (optional).

ProgIf Programming interface (optional).

Driver Kernel driver currently handling the device (optional, Linux only).

Module Kernel module reporting that it is capable of handling the device (optional, Linux only).

New tags can be added in future versions, so you should silently ignore any tags you don’t recognize.

Backward-compatible verbose format (-vm)
In this mode, lspci tries to be perfectly compatible with its old versions. It’s almost the same as the regular verbose format, but the Device tag is used for both the slot and the device name, so it occurs twice in a single record. Please avoid using this format in any new code.


A list of all known PCI ID’s (vendors, devices, classes and subclasses). Maintained at, use the update-pciids utility to download the most recent version.

If lspci is compiled with support for compression, this file is tried before pci.ids.

An interface to PCI bus configuration space provided by the post-2.1.82 Linux kernels. Contains per-bus subdirectories with per-card config space files and a devices file containing a list of all PCI devices.


Sometimes, lspci is not able to decode the configuration registers completely. This usually happens when not enough documentation was available to the authors. In such cases, it at least prints the <?> mark to signal that there is potentially something more to say. If you know the details, patches will be of course welcome.

Access to the extended configuration space is currently supported only by the linux_sysfs back-end.

See Also

setpci(8) , update-pciids(8)


The PCI Utilities are maintained by Martin Mares <>.

Table of Contents

use open software!