Song Liu
Software Engineer at Meta
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSong Liu is a Linux Kernel software engineer with 11 years of experience, currently working at Meta in Mountain View and long-tenured at Facebook. He maintains and reviews BPF (tracing/general) and md/raid subsystems, bringing deep kernel internals expertise to production tracing and storage features. His open-source work on libbpf and contributions to fio demonstrate a focus on observability and I/O performance—adding practical features like richer BPF program info, task-stack dumping, and realistic rate-limiting simulations. A PhD-trained engineer from Northwestern with an undergraduate degree from Tsinghua, he blends academic rigor with hands-on systems development at scale. Notably, his contributions span both kernel-maintainer responsibilities and userland tooling that tighten the feedback loop between tracing instrumentation and performance benchmarking.
11 years of coding experience
1 year of employment as a software developer
PhD, Computer Engineering, PhD, Computer Engineering at Northwestern UniversityBachelor, Electronic Engineering, Bachelor, Electronic Engineering at Tsinghua UniversityShijiazhuang second middle school
Github Skills (24)
raid10
benchmark10
performance-analytics10
performance-monitor10
rate-limiting10
system-programming10
benchmarking10
c1110
linux-kernel10
bpf10
data-structure10
performance-measurement10
c1710
performance-analysis10
performance-tuning10
Programming languages (6)
Github contributions (5)
libbpf/libbpf
Mar 2019 - Nov 2021
Automated upstream mirror for libbpf stand-alone build.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 reviews, 12 commits, 8 PRs in 2 years 8 months
Contributions summary:Song primarily contributes to the `libbpf` library, focusing on enhancing its functionality and usability. Their work includes implementing new features, such as `bpf_program__get_prog_info_linear()`, which streamlines the retrieval of BPF program information. They also added support for BPF_ENABLE_STATS and introduced helper functions to dump task stacks, enabling richer tracing capabilities. Their contributions appear to involve extending the BPF functionalities and improving its interaction with the kernel.
bpfstand-alonestandtracingebpf
axboe/fio
Nov 2015 - May 2020
Flexible I/O Tester
Role in this project:Back-end Developer
Back-end DeveloperContributions:7 commits, 5 PRs, 4 comments in 4 years 7 months
Contributions summary:Song contributed to the core functionality of the `fio` utility, focusing on performance and rate limiting features. They implemented a Poisson process for request flow simulation, enhancing the realism of rate-limited benchmarks. Further contributions include fixing an issue in the random number generation and refactoring btrace2fio to generate separate jobs for combined read/write and trim operations. Additional commits addressed unit discrepancies in latency calculations and introduced an option to maintain latency targets during benchmark runs.
testingflexible
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.