1 =============
2 BPF licensing
3 =============
4
5 Background
6 ==========
7
8 * Classic BPF was BSD licensed
9
10 "BPF" was originally introduced as BSD Packet
11 http://www.tcpdump.org/papers/bpf-usenix93.pdf
12 set and its implementation came from BSD with
13 instruction set is now known as "classic BPF".
14
15 However an instruction set is a specification
16 similar to a programming language. It is not
17 application of a BSD license may be misleading
18 instruction set may enjoy no copyright protect
19
20 * eBPF (extended BPF) instruction set continue
21
22 In 2014, the classic BPF instruction set was s
23 typically refer to this instruction set as eBP
24 The eBPF instruction set is still BSD licensed
25
26 Implementations of eBPF
27 =======================
28
29 Using the eBPF instruction set requires implem
30 and user space.
31
32 In Linux Kernel
33 ---------------
34
35 The reference implementations of the eBPF inte
36 compilers are part of Linux and are GPLv2 lice
37 eBPF helper functions is also GPLv2 licensed.
38 and verifiers are called eBPF runtime.
39
40 In User Space
41 -------------
42
43 There are also implementations of eBPF runtime
44 functions) under
45 Apache2 (https://github.com/iovisor/ubpf),
46 MIT (https://github.com/qmonnet/rbpf), and
47 BSD (https://github.com/DPDK/dpdk/blob/main/li
48
49 In HW
50 -----
51
52 The HW can choose to execute eBPF instruction
53 in HW or via the use of implementing firmware
54
55 In other operating systems
56 --------------------------
57
58 Other kernels or user space implementations of
59 can have proprietary licenses.
60
61 Using BPF programs in the Linux kernel
62 ======================================
63
64 Linux Kernel (while being GPLv2) allows linkin
65 under these rules:
66 Documentation/process/license-rules.rst
67
68 When a kernel module is loaded, the linux kern
69 intends to use. If any function is marked as "
70 module or program has to have GPL compatible l
71
72 Loading BPF program into the Linux kernel is s
73 module. BPF is loaded at run time and not stat
74 kernel. BPF program loading follows the same l
75 modules. BPF programs can be proprietary if th
76 helper functions.
77
78 Further, some BPF program types - Linux Securi
79 Congestion Control (struct_ops), as of Aug 202
80 compatible even if they don't use "GPL only" h
81 registration step of LSM and TCP congestion co
82 kernel is done through EXPORT_SYMBOL_GPL kerne
83 and struct_ops BPF programs are implicitly cal
84 The same restriction applies to BPF programs t
85 directly via unstable interface also known as
86
87 Packaging BPF programs with user space applica
88 ==============================================
89
90 Generally, proprietary-licensed applications a
91 written for the Linux kernel in the same packa
92 separate executable processes. This applies to
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.