OpenBSD Journal

Paul Irofti: t2k13: Hibernating on the Loongson

Contributed by pitrh on from the sleep deep dept.

Further swelling our t2k13 hackathon coverage, Paul Irofti (pirofti@) sent us this report:
I spent most of the hackathon adding hibernate support for the Loongson architecture.

I found the work that mlarkin@ and deraadt@ have done on hibernate to be quite interesting. Unfortunately during their initial work on i386 and amd64 I was focused on different topics and I kind of lost the train on that one.

No matter, I decided to join in on the hibernate effort and Loongson seemed like the perfect architecture for that. I already added suspend support for it a few years ago so hibernation would be the next logical step.

The Lemote is a very good port for testing the hibernate MI framework as it's very different from the currently supported ports which will help spot bugs and other issues.

I started reading a lot of code and documentation about the MIPS64 processor and the Loongson memory layout. Soon enough I had a kernel that hooked up the machine independent hibernate framework into the machine dependent bits of my Lemote.

I slowly started replacing the stubs and writing proper functionality for the machine dependent bits while keeping an eye on the hibernate framework. I talked a lot with mlarkin@ mostly asking him for advice or requesting small changes that would make the MI framework even more MI.

At the same time I worked with miod@, making sure my approach fits the mips64 ports philosophy and that my code is not too disgusting.

While hibernate support is far from ready, the initial bits are in and more work is on the way.

Besides hibernate I also had fun helping jasper@ with adding octeon MP support and hacking a second stage bootloader. It got me kind of interested in that port, too bad getting a board is so complicated.

I also did a small bit of ports work, making sure my ports are up to date and still working. I had the pleasure to discover that minitube works so much better nowadays which I guess it's due to kms support on the intel boards.

So all in all the hackathon was very productive for me and I had a lot of fun working on OpenBSD and wandering around Toronto. I would like to thank our organizer (and babysitter at times) Ken Westerback for the huge effort he put into making this an amazing hackathon.

Thank you so much for that Ken! Please do hold another one next year and in the years to come!

Thank you, Paul, for excellent work and this report!

(Comments are closed)


Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]