https://github.com/wenxin-wang/flowlabel-switching

flowlabel-base label switching for IPv6
https://github.com/wenxin-wang/flowlabel-switching

Last synced: 2 months ago
JSON representation

flowlabel-base label switching for IPv6

• Host: GitHub
• URL: https://github.com/wenxin-wang/flowlabel-switching
• Owner: wenxin-wang
• License: mit
• Archived: true
• Created: 2019-01-20T15:47:46.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2019-05-23T04:47:57.000Z (over 5 years ago)
• Last Synced: 2024-08-02T01:20:03.434Z (5 months ago)
• Language: C
• Size: 60.5 KB
• Stars: 2
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Edge

1. labeling: lwt xmit

2. switching: like backbone

3. unlabeling: lwt xmit

## Labeling

lwt xmit bpf cannot have arguments

1. lpm match for a label value

2. if not match, unset flowlabel

3. if match, change flowlabel

4. continue routing

## Unlabeling

Just set flowlabel to 0

# Backbone

xdp bpf

## Fallback

unlabel + routing: switching fallback to this when seeing 0-label or unknown

label (sets label to 0 before routing)

# Build

## Prerequisites

1. linux-kernel >= 4.18 (for `fib_lookup`): See [BPF Features by Linux Kernel

   Version](https://github.com/iovisor/bcc/blob/master/docs/kernel-versions.md)

2. bcc: for libbpf

## Build & Install

```bash

cd flowlabel-switching/

make

```

# Examples

See [test/]().

# About `ip netns exec`

The default bpf filesystem, used for pinning bpf programs and maps, is mounted

on `/sys/fs/bpf/` by `iproute2` programs. However, `ip netns exec` happens to

unmount `/sys` before `exec`, so the called program never sees that bpf

filesystem.

And it seems that bpf filesystem cannot be mounted by any command invoked with

`ip netns exec`. I don't know the reason behind.

So before calling `ip netns exec`, mount a bpf filesystem somewhere other than

`/sys`, e.g. `/run/flsw/bpffs/`, and `ip netns exec` will see that filesystem.

And by setting `TC_BPF_MNT` environment variable to ~~subdirectories of the

previously mounted bpf filesystem.~~ (iproute2 tries to mount bpf filesystem

whenever `TC_BPF_MNT` is not a bpf mount point, whether it's a subdirectory of a

bpf fs or not, probably with good reason) different mount points of bpf

filesystems, we can have different bpf objects with the same name for different

netns.