Updates tcp_proxy to use an AF_PACKET and veth devices.

tcp_proxy now uses an AF_PACKET socket as the FD for netstack link layer
endpoint instead of a tap device. It also changes the link layer endpoint to use
PacketMMap dispatch instead of Readv. This reduces overall cpu and reflects the
current runsc setup which uses PacketMMap and also uses veth devices to receive
packets.

Also fixed a bug in gonet where Read() was not doing coalescing read and would
read small amounts at a time.

PiperOrigin-RevId: 234714768
1 file changed
tree: 409a0ca560561f246d63ac418c0b6c5beb1e0427
  1. dhcp/
  2. gate/
  3. ilist/
  4. rand/
  5. sleep/
  6. tcpip/
  7. tmutex/
  8. waiter/
  9. AUTHORS
  10. CONTRIBUTING.md
  11. CONTRIBUTORS
  12. LICENSE
  13. README.md
README.md

Netstack

Netstack is a network stack written in Go.

Getting started

Try it out on Linux by installing the tun_tcp_echo demo:

go install github.com/google/netstack/tcpip/sample/tun_tcp_echo

Create a TUN device with:

[sudo] ip tuntap add user <username> mode tun <device-name>
[sudo] ip link set <device-name> up
[sudo] ip addr add <ipv4-address>/<mask-length> dev <device-name>

Then run with:

tun_tcp_echo <device-name> <ipv4-address> <port>

Contributions

Please see CONTRIBUTING.md for more details.

Disclaimer

This is not an official Google product (experimental or otherwise), it is just code that happens to be owned by Google.