Re: Introduction and project intent: Emulating missing linux syscalls
Stephen Borrill <[email protected]> Wed, 3 Jun 2026 10:07:02 +0100 (BST)
| Newsgroups | gmane.os.netbsd.devel.kernel |
|---|---|
| Message-ID | <Pine.NEB.4.64.2606030959580.2896@ugly> |
On Tue, 2 Jun 2026, Ilia Javan wrote: > Hello tech-kern, > > My name is Ilia, and I am an undergraduate computer engineering > student. I am writing to introduce myself and declare my intent to > spend my summer working on adding missing Linux syscalls to the > compat_linux translation layer. While this is a personal project and > not affiliated with Google Summer of Code, I plan to treat the time > commitment and deliverables as seriously as though it were. Hi Ilia, There is currently a GSoC project in place relating to Linux emulation. It is focussing on testing the framework and it is anticipated that this will highlight broken/missing syscalls. There are syscalls that we know will need to be added (based on a previous GSoC) and the testing will help with that. Basically, what I'm saying is that I think it would be better to wait until GSoC has completed as this will give plenty of tasks that need to be done. > My background includes C programming and CPU hardware design on FPGAs, > but this will be my first non-trivial contribution to an open-source > project and my first dive into kernel development. > > To prepare, I have spent the past few weeks reading through texts on > operating system fundamentals (Advanced Programming in the UNIX > Environment, The Design and Implementation of the 4.4BSD Operating > System, and Understanding the Linux Kernel). To get hands-on > experience with the NetBSD ecosystem, CVS, and the GNATS system, I > recently submitted a patch for a minor pluralization bug in the > shutdown(8) utility (PR bin/60248). Going through the submission and > getting feedback on it from the developers has been a great way to > understand the workflow. > > My roadmap for the immediate future includes: > > - Finding the missing syscalls: I plan to start by investigating the > y-cruncher binary, as reported in PR 59589 (Linux emulation fails on > y-cruncher). I will trace its execution on my amd64 NetBSD QEMU > environment to identify exactly which syscalls or features it is > missing. I will also test other Linux binaries to map out additional > gaps. > > - Finalizing a timeline: Once I have traced the missing syscalls for > y-cruncher (or another suitable binary), I will share a detailed > project timeline for the remaining 3 months of my summer. > > I am very open to suggestions if the community has a specific Linux > binary or missing syscall they would prioritize for compatibility over > y-cruncher. > > Thank you for your time. I am eagerly looking forward to any advice, > feedback, or suggested reading materials as I begin this project. > > Best regards, > > Ilia >