1 # SPDX-License-Identifier: GPL-2.0-only
3 # Network device configuration
9 bool "Network device support"
11 You can say N here if you don't intend to connect your Linux box to
12 any other computer at all.
14 You'll have to say Y if your computer contains a network card that
15 you want to use under Linux. If you are going to run SLIP or PPP over
16 telephone line or null modem cable you need say Y here. Connecting
17 two machines with parallel ports using PLIP needs this, as well as
18 AX.25/KISS for sending Internet traffic over amateur radio links.
20 See also "The Linux Network Administrator's Guide" by Olaf Kirch and
21 Terry Dawson. Available at <http://www.tldp.org/guides.html>.
25 # All the following symbols are dependent on NETDEVICES - do not repeat
26 # that for each of the symbols.
34 bool "Network core driver support"
36 You can say N here if you do not intend to use any of the
37 networking core drivers (i.e. VLAN, bridging, bonding, etc.)
42 tristate "Bonding driver support"
44 depends on IPV6 || IPV6=n
45 depends on TLS || TLS_DEVICE=n
47 Say 'Y' or 'M' if you wish to be able to 'bond' multiple Ethernet
48 Channels together. This is called 'Etherchannel' by Cisco,
49 'Trunking' by Sun, 802.3ad by the IEEE, and 'Bonding' in Linux.
51 The driver supports multiple bonding modes to allow for both high
52 performance and high availability operation.
54 Refer to <file:Documentation/networking/bonding.rst> for more
57 To compile this driver as a module, choose M here: the module
58 will be called bonding.
61 tristate "Dummy net driver support"
63 This is essentially a bit-bucket device (i.e. traffic you send to
64 this device is consigned into oblivion) with a configurable IP
65 address. It is most commonly used in order to make your currently
66 inactive SLIP address seem like a real address for local programs.
67 If you use SLIP or PPP, you might want to say Y here. It won't
68 enlarge your kernel. What a deal. Read about it in the Network
69 Administrator's Guide, available from
70 <http://www.tldp.org/docs.html#guide>.
72 To compile this driver as a module, choose M here: the module
76 tristate "WireGuard secure network tunnel"
77 depends on NET && INET
78 depends on IPV6 || !IPV6
82 select CRYPTO_LIB_CURVE25519
83 select CRYPTO_LIB_CHACHA20POLY1305
84 select CRYPTO_LIB_BLAKE2S
85 select CRYPTO_CHACHA20_X86_64 if X86 && 64BIT
86 select CRYPTO_POLY1305_X86_64 if X86 && 64BIT
87 select CRYPTO_BLAKE2S_X86 if X86 && 64BIT
88 select CRYPTO_CURVE25519_X86 if X86 && 64BIT
89 select ARM_CRYPTO if ARM
90 select ARM64_CRYPTO if ARM64
91 select CRYPTO_CHACHA20_NEON if (ARM || ARM64) && KERNEL_MODE_NEON
92 select CRYPTO_POLY1305_NEON if ARM64 && KERNEL_MODE_NEON
93 select CRYPTO_POLY1305_ARM if ARM
94 select CRYPTO_BLAKE2S_ARM if ARM
95 select CRYPTO_CURVE25519_NEON if ARM && KERNEL_MODE_NEON
96 select CRYPTO_CHACHA_MIPS if CPU_MIPS32_R2
97 select CRYPTO_POLY1305_MIPS if CPU_MIPS32 || (CPU_MIPS64 && 64BIT)
99 WireGuard is a secure, fast, and easy to use replacement for IPSec
100 that uses modern cryptography and clever networking tricks. It's
101 designed to be fairly general purpose and abstract enough to fit most
102 use cases, while at the same time remaining extremely simple to
103 configure. See www.wireguard.com for more info.
105 It's safe to say Y or M here, as the driver is very lightweight and
106 is only in use when an administrator chooses to add an interface.
108 config WIREGUARD_DEBUG
109 bool "Debugging checks and verbose messages"
112 This will write log messages for handshake and other events
113 that occur for a WireGuard interface. It will also perform some
114 extra validation checks and unit tests at various points. This is
115 only useful for debugging.
117 Say N here unless you know what you're doing.
120 tristate "EQL (serial line load balancing) support"
122 If you have two serial connections to some other computer (this
123 usually requires two modems and two telephone lines) and you use
124 SLIP (the protocol for sending Internet traffic over telephone
125 lines) or PPP (a better SLIP) on them, you can make them behave like
126 one double speed connection using this driver. Naturally, this has
127 to be supported at the other end as well, either with a similar EQL
128 Linux driver or with a Livingston Portmaster 2e.
130 Say Y if you want this and read
131 <file:Documentation/networking/eql.rst>. You may also want to read
132 section 6.2 of the NET-3-HOWTO, available from
133 <http://www.tldp.org/docs.html#howto>.
135 To compile this driver as a module, choose M here: the module
136 will be called eql. If unsure, say N.
139 bool "Fibre Channel driver support"
140 depends on SCSI && PCI
142 Fibre Channel is a high speed serial protocol mainly used to connect
143 large storage devices to the computer; it is compatible with and
144 intended to replace SCSI.
146 If you intend to use Fibre Channel, you need to have a Fibre channel
147 adaptor card in your computer; say Y here and to the driver for your
148 adaptor below. You also should have said Y to "SCSI support" and
149 "SCSI generic support".
152 tristate "Intermediate Functional Block support"
153 depends on NET_CLS_ACT
156 This is an intermediate driver that allows sharing of
158 To compile this driver as a module, choose M here: the module
159 will be called ifb. If you want to use more than one ifb
160 device at a time, you need to compile this driver as a module.
161 Instead of 'ifb', the devices will then be called 'ifb0',
163 Look at the iproute2 documentation directory for usage etc
165 source "drivers/net/team/Kconfig"
168 tristate "MAC-VLAN support"
170 This allows one to create virtual interfaces that map packets to
171 or from specific MAC addresses to a particular interface.
173 Macvlan devices can be added using the "ip" command from the
174 iproute2 package starting with the iproute2-2.6.23 release:
176 "ip link add link <real dev> [ address MAC ] [ NAME ] type macvlan"
178 To compile this driver as a module, choose M here: the module
179 will be called macvlan.
182 tristate "MAC-VLAN based tap driver"
187 This adds a specialized tap character device driver that is based
188 on the MAC-VLAN network interface, called macvtap. A macvtap device
189 can be added in the same way as a macvlan device, using 'type
190 macvtap', and then be accessed through the tap user space interface.
192 To compile this driver as a module, choose M here: the module
193 will be called macvtap.
199 select NET_L3_MASTER_DEV
202 tristate "IP-VLAN support"
204 depends on IPV6 || !IPV6
206 This allows one to create virtual devices off of a main interface
207 and packets will be delivered based on the dest L3 (IPv6/IPv4 addr)
208 on packets. All interfaces (including the main interface) share L2
209 making it transparent to the connected L2 switch.
211 Ipvlan devices can be added using the "ip" command from the
212 iproute2 package starting with the iproute2-3.19 release:
214 "ip link add link <main-dev> [ NAME ] type ipvlan"
216 To compile this driver as a module, choose M here: the module
217 will be called ipvlan.
220 tristate "IP-VLAN based tap driver"
225 This adds a specialized tap character device driver that is based
226 on the IP-VLAN network interface, called ipvtap. An ipvtap device
227 can be added in the same way as a ipvlan device, using 'type
228 ipvtap', and then be accessed through the tap user space interface.
230 To compile this driver as a module, choose M here: the module
231 will be called ipvtap.
234 tristate "Virtual eXtensible Local Area Network (VXLAN)"
236 select NET_UDP_TUNNEL
239 This allows one to create vxlan virtual interfaces that provide
240 Layer 2 Networks over Layer 3 Networks. VXLAN is often used
241 to tunnel virtual network infrastructure in virtualized environments.
242 For more information see:
243 http://tools.ietf.org/html/draft-mahalingam-dutt-dcops-vxlan-02
245 To compile this driver as a module, choose M here: the module
246 will be called vxlan.
249 tristate "Generic Network Virtualization Encapsulation"
251 depends on IPV6 || !IPV6
252 select NET_UDP_TUNNEL
255 This allows one to create geneve virtual interfaces that provide
256 Layer 2 Networks over Layer 3 Networks. GENEVE is often used
257 to tunnel virtual network infrastructure in virtualized environments.
258 For more information see:
259 http://tools.ietf.org/html/draft-gross-geneve-02
261 To compile this driver as a module, choose M here: the module
262 will be called geneve.
265 tristate "Bare UDP Encapsulation"
267 depends on IPV6 || !IPV6
268 select NET_UDP_TUNNEL
271 This adds a bare UDP tunnel module for tunnelling different
272 kinds of traffic like MPLS, IP, etc. inside a UDP tunnel.
274 To compile this driver as a module, choose M here: the module
275 will be called bareudp.
278 tristate "GPRS Tunneling Protocol datapath (GTP-U)"
280 select NET_UDP_TUNNEL
282 This allows one to create gtp virtual interfaces that provide
283 the GPRS Tunneling Protocol datapath (GTP-U). This tunneling protocol
284 is used to prevent subscribers from accessing mobile carrier core
285 network infrastructure. This driver requires a userspace software that
286 implements the signaling protocol (GTP-C) to update its PDP context
287 base, such as OpenGGSN <http://git.osmocom.org/openggsn/). This
288 tunneling protocol is implemented according to the GSM TS 09.60 and
289 3GPP TS 29.060 standards.
291 To compile this drivers as a module, choose M here: the module
295 tristate "IEEE 802.1AE MAC-level encryption (MACsec)"
301 MACsec is an encryption standard for Ethernet.
304 tristate "Network console logging support"
306 If you want to log kernel messages over the network, enable this.
307 See <file:Documentation/networking/netconsole.rst> for details.
309 config NETCONSOLE_DYNAMIC
310 bool "Dynamic reconfiguration of logging targets"
311 depends on NETCONSOLE && SYSFS && CONFIGFS_FS && \
312 !(NETCONSOLE=y && CONFIGFS_FS=m)
314 This option enables the ability to dynamically reconfigure target
315 parameters (interface, IP addresses, port numbers, MAC addresses)
316 at runtime through a userspace interface exported using configfs.
317 See <file:Documentation/networking/netconsole.rst> for details.
323 config NET_POLL_CONTROLLER
327 tristate "Virtual Ethernet over NTB Transport"
328 depends on NTB_TRANSPORT
331 tristate "RapidIO Ethernet over messaging driver support"
334 config RIONET_TX_SIZE
335 int "Number of outbound queue entries"
339 config RIONET_RX_SIZE
340 int "Number of inbound queue entries"
345 tristate "Universal TUN/TAP device driver support"
349 TUN/TAP provides packet reception and transmission for user space
350 programs. It can be viewed as a simple Point-to-Point or Ethernet
351 device, which instead of receiving packets from a physical media,
352 receives them from user space program and instead of sending packets
353 via physical media writes them to the user space program.
355 When a program opens /dev/net/tun, driver creates and registers
356 corresponding net device tunX or tapX. After a program closed above
357 devices, driver will automatically delete tunXX or tapXX device and
358 all routes corresponding to it.
360 Please read <file:Documentation/networking/tuntap.rst> for more
363 To compile this driver as a module, choose M here: the module
366 If you don't know what to use this for, you don't need it.
371 This option is selected by any driver implementing tap user space
372 interface for a virtual interface to re-use core tap functionality.
374 config TUN_VNET_CROSS_LE
375 bool "Support for cross-endian vnet headers on little-endian kernels"
378 This option allows TUN/TAP and MACVTAP device drivers in a
379 little-endian kernel to parse vnet headers that come from a
380 big-endian legacy virtio device.
382 Userspace programs can control the feature using the TUNSETVNETBE
383 and TUNGETVNETBE ioctls.
385 Unless you have a little-endian system hosting a big-endian virtual
386 machine with a legacy virtio NIC, you should say N.
389 tristate "Virtual ethernet pair device"
391 This device is a local ethernet tunnel. Devices are created in pairs.
392 When one end receives the packet it appears on its pair and vice
396 tristate "Virtio network driver"
400 This is the virtual network driver for virtio. It can be used with
401 QEMU based VMMs (like KVM or Xen). Say Y or M.
404 tristate "Virtual netlink monitoring device"
406 This option enables a monitoring net device for netlink skbs. The
407 purpose of this is to analyze netlink messages with packet sockets.
408 Thus applications like tcpdump will be able to see local netlink
409 messages if they tap into the netlink device, record pcaps for further
410 diagnostics, etc. This is mostly intended for developers or support
411 to debug netlink issues. If unsure, say N.
414 tristate "Virtual Routing and Forwarding (Lite)"
415 depends on IP_MULTIPLE_TABLES
416 depends on NET_L3_MASTER_DEV
417 depends on IPV6 || IPV6=n
418 depends on IPV6_MULTIPLE_TABLES || IPV6=n
420 This option enables the support for mapping interfaces into VRF's. The
421 support enables VRF devices.
424 tristate "Virtual vsock monitoring device"
425 depends on VHOST_VSOCK
427 This option enables a monitoring net device for vsock sockets. It is
428 mostly intended for developers or support to debug vsock issues. If
432 tristate "MHI network driver"
435 This is the network driver for MHI bus. It can be used with
436 QCOM based WWAN modems (like SDX55). Say Y or M.
443 source "drivers/net/arcnet/Kconfig"
445 source "drivers/atm/Kconfig"
447 source "drivers/net/caif/Kconfig"
449 source "drivers/net/dsa/Kconfig"
451 source "drivers/net/ethernet/Kconfig"
453 source "drivers/net/fddi/Kconfig"
455 source "drivers/net/hippi/Kconfig"
457 source "drivers/net/ipa/Kconfig"
460 tristate "General Instruments Surfboard 1000"
463 This is a driver for the General Instrument (also known as
464 NextLevel) SURFboard 1000 internal
465 cable modem. This is an ISA card which is used by a number of cable
466 TV companies to provide cable modem access. It's a one-way
467 downstream-only cable modem, meaning that your upstream net link is
468 provided by your regular phone modem.
470 At present this driver only compiles as a module, so say M here if
471 you have this card. The module will be called sb1000. Then read
472 <file:Documentation/networking/device_drivers/cable/sb1000.rst> for
473 information on how to use this module, as it needs special ppp
474 scripts for establishing a connection. Further documentation
475 and the necessary scripts can be found at:
477 <http://www.jacksonville.net/~fventuri/>
478 <http://home.adelphia.net/~siglercm/sb1000.html>
479 <http://linuxpower.cx/~cable/>
481 If you don't have this card, of course say N.
483 source "drivers/net/phy/Kconfig"
485 source "drivers/net/mdio/Kconfig"
487 source "drivers/net/pcs/Kconfig"
489 source "drivers/net/plip/Kconfig"
491 source "drivers/net/ppp/Kconfig"
493 source "drivers/net/slip/Kconfig"
495 source "drivers/s390/net/Kconfig"
497 source "drivers/net/usb/Kconfig"
499 source "drivers/net/wireless/Kconfig"
501 source "drivers/net/wan/Kconfig"
503 source "drivers/net/ieee802154/Kconfig"
505 config XEN_NETDEV_FRONTEND
506 tristate "Xen network device frontend driver"
508 select XEN_XENBUS_FRONTEND
512 This driver provides support for Xen paravirtual network
513 devices exported by a Xen network driver domain (often
516 The corresponding Linux backend driver is enabled by the
517 CONFIG_XEN_NETDEV_BACKEND option.
519 If you are compiling a kernel for use as Xen guest, you
520 should say Y here. To compile this driver as a module, chose
521 M here: the module will be called xen-netfront.
523 config XEN_NETDEV_BACKEND
524 tristate "Xen backend network device"
525 depends on XEN_BACKEND
527 This driver allows the kernel to act as a Xen network driver
528 domain which exports paravirtual network devices to other
529 Xen domains. These devices can be accessed by any operating
530 system that implements a compatible front end.
532 The corresponding Linux frontend driver is enabled by the
533 CONFIG_XEN_NETDEV_FRONTEND configuration option.
535 The backend driver presents a standard network device
536 endpoint for each paravirtual network device to the driver
537 domain network stack. These can then be bridged or routed
538 etc in order to provide full network connectivity.
540 If you are compiling a kernel to run in a Xen network driver
541 domain (often this is domain 0) you should say Y here. To
542 compile this driver as a module, chose M here: the module
543 will be called xen-netback.
546 tristate "VMware VMXNET3 ethernet driver"
547 depends on PCI && INET
548 depends on !(PAGE_SIZE_64KB || ARM64_64K_PAGES || \
549 IA64_PAGE_SIZE_64KB || MICROBLAZE_64K_PAGES || \
550 PARISC_PAGE_SIZE_64KB || PPC_64K_PAGES)
552 This driver supports VMware's vmxnet3 virtual ethernet NIC.
553 To compile this driver as a module, choose M here: the
554 module will be called vmxnet3.
557 tristate "FUJITSU Extended Socket Network Device driver"
560 This driver provides support for Extended Socket network device
561 on Extended Partitioning of FUJITSU PRIMEQUEST 2000 E2 series.
564 tristate "Networking over USB4 and Thunderbolt cables"
565 depends on USB4 && INET
567 Select this if you want to create network between two computers
568 over a USB4 and Thunderbolt cables. The driver supports Apple
569 ThunderboltIP protocol and allows communication with any host
570 supporting the same protocol including Windows and macOS.
572 To compile this driver a module, choose M here. The module will be
573 called thunderbolt-net.
575 source "drivers/net/hyperv/Kconfig"
578 tristate "Simulated networking device"
581 depends on IPV6 || IPV6=n
584 This driver is a developer testing tool and software model that can
585 be used to test various control path networking APIs, especially
588 To compile this driver as a module, choose M here: the module
589 will be called netdevsim.
592 tristate "Failover driver"
595 This provides an automated failover mechanism via APIs to create
596 and destroy a failover master netdev and manages a primary and
597 standby slave netdevs that get registered via the generic failover
598 infrastructure. This can be used by paravirtual drivers to enable
599 an alternate low latency datapath. It also enables live migration of
600 a VM with direct attached VF by failing over to the paravirtual
601 datapath when the VF is unplugged.