Description
Many security tools need to record system calls like execve. Using the Linux audit system for this can have a detrimental performance impact in some cases.
Goals
The goal is to investigate eBPF as an alternative and do some benchmarking to see the impact and how it compares to using the audit subsystem.
Progress
BPF done - traceexec
Resources
This project is part of:
Hack Week 24
Activity
Comments
-
about 1 year ago by jiriwiesner | Reply
If I were to do this task the syscount script from bcc-tools would be my starting point: https://github.com/iovisor/bcc/blob/master/tools/syscount.py
Similar Projects
bpftrace contribution by mkoutny
Description
bpftrace is a great tool, no need to sing odes to it here. It can access any kernel data and process them in real time. It provides helpers for some common Linux kernel structures but not all.
Goals
- set up bpftrace toolchain
- learn about bpftrace implementation and internals
- implement support for
percpu_counters - look into some of the first issues
- send a refined PR (on Thu)
Resources
eBPF bytecode emitter in Haskell by kalfalakh
Description
Newbie level knowledge of eBPF and some knowledge of Haskell. The goal for this hackweek is to catch two birds with one stone; get familiar with eBPF and learn more about Haskell, hence implement something related to eBPF in Haskell. Given an input, which is a program, represented as eBPF instructions, prepare a fully built eBPF bytecode ready to be loaded into the Kernel.
Goals
- Recap on ADTs in Haskell, type classes and their instances / deriving, on pattern matching and higher order functions
- Read and understand RFC 9669
- Implement the entire pipeline; parsing the input, verifying the input, building instructions, encoding them and concatenating them
- Deal with all the learning, bugs and difficulties encountered on the way
Progress
Last day of hackweek and this is what I have so far.
- ADTs implemented to correctly type all operations, classes, opcodes, instructions and other instruction specific types
- opcodes and opcode builders based on specification are implemented.
- Encoder is ready for 64-bit instruction format. Extended 128-bit instruction format is not yet supported
- Only ALU and ALU64 instructions API is provided. JMP, LD, ST for base32 and base64 are missing. swap and atomic are missing
- All ALU and ALU64 instructions (except MOV and MOVSX) are tested manually and are correctly encoded
Whats next?
There is still a lot left to do:
- Input parser and formatter + some kind of basic verifier
- API for remaining instruction classes
- Proper testing mechanism
- swap and atomic instructions
- Loader
Resources
- RFC 9669
- https://ebpf.io/
- https://www.haskell.org/documentation/