Home News Featured Blogs Libby Clark Best Quotes from the Linux Kernel Developer Panel

Best Quotes from the Linux Kernel Developer Panel

Kernel panelLinux kernel developers Greg Kroah-Hartman, Jens Axboe, Dave Chinner, Matthew Garrett, and Mel Gorman participated in a panel discussion, moderated by LWN Editor Jon Corbet, at Collaboration Summit on Wednesday. Here are some of the highlights. For the full session, view the video, below.

For more information on what each developer is working on now, see our Linux kernel panel preview

On why companies hire kernel developers:

“We're finding out more and more there's actually a commercial imperative to give your software away to help others, because they help you and then everyone benefits... This is the way to do business.” - Dave Chinner, Red Hat.

“As Facebook has demonstrated with their Open Compute platform initiative, it actually makes economic sense to develop this in the open and not just improve your own data center but also improve data centers elsewhere … I view this as pretty much the same thing, just on the software side.” - Jens Axboe, Facebook.

On Facebook's participation in kernel development:

Jon Corbet: “So do we need a certain number of likes to get a block patch merged now?”
Jens Axboe: “That's the intent. I'm setting up a review group on Facebook now.”

On working with developers at competing companies:

“For the most part there's very good collaboration between companies... and these people are my friends... Even when someone is working for competitor companies it's hard to think of them that way.” - Matthew Garrett, Nebula.

“Even though the other companies are competitors, there's also a symbiotic relationship. It reached the point a GregKHlong time ago that there's no single company that has access to enough expertise in all areas to completely go it alone.” - Mel Gorman, SUSE.

On their relationship with userspace developers:

“There are some cases when userspace developers come to us and say, 'this kernel behavior is causing problems,' and our immediate reaction is 'no, the kernel is fine, you're using it wrong.' And sometimes that's true and sometimes in fact we've created a user interface that's impossible to interact with. But we tend to lean a bit too far to the side that userspace programmers know less about the kernel than we do, which is probably true, and therefore userspace programmers are usually wrong which may not be true. It would be nice if we didn't always jump to that conclusion quite so quickly.” - Matthew Garrett.

On the pace of change within the kernel:

“We are blazing new trails. We've been doing that for a long time. We're way past what Unix could ever do and that's hard to do.” Greg Kroah-Hartman, The Linux Foundation.

“The Linux kernel is one of the largest collaborative software projects in the history of the world and has almost nothing in the way of formalized management structure. We have people who have a strong operating systems background who have been contributing code, and then we have people like me. I have a background in fruit fly genetics and yet someone lets me get close to the Linux kernel; this seems wrong. And then we have people who are genuinely kids in their bedroom. It's a miracle it works as well as it does. We should be astonished that we're able to get it so right so much of the time.” - Matthew Garrett.

Mel-Gorman“We still do it better than anybody else ever has, so give us credit for that.” Greg Kroah-Hartman.

On bug fixing and automated testing:

“We're less reliant on eyeballs to find hard-to-find bugs – they just don't get out of a developer's machine because they're found (in testing) before review is even considered. The bar has gone up and the code coming out from the developers has improved; it's better quality. We still need better test coverage, (but) it's better than it was.” - Dave Chinner.

Greg Kroah-Hartman: The tests we have, even in the kernel tree don't work.
Mel Gorman: We need tests for the tests.
Greg Kroah-Hartman: No we need tests that work.

On UEFI secure boot and Matthew Garrett's recent award from the Free Software Foundation:

Corbet: How many of you are using systems with secure boot enabled now? (A few people raise their hands.)
Audience member: It's the first thing I turn off on a new machine.
Garett : You could just try it, except I guess that most of you probably haven't bought new hardware for years because you're kernel developers.
Audience member: I'm on my fifth motherboard thanks to UEFI, so...
Corbet: Well the three of you who are using it can thank Matthew.



Subscribe to Comments Feed
  • Linux Said:

    Learning a new skill can be a challenge, especially when the topic seems so vast.

  • Jonathan Hartley Said:

    I know I'm not going to teach the kernel devs anything here, but for those, like me, reading along... > Mel Gorman: We need tests for the tests "tests for the tests" are the TDD process: Write a test that the code-under-test does X, run the test to see it fail, add (or modify) the code-under-test to do X, and then run the test again to see it pass, At the individual developer's discretion, you might then choose to also trivially modify the code-under-test to do Y (something similar to X, but not quite identical.) Run the test again to watch it fail, then undo that modification. This procedure acts as a manual verification that your test really is correct. It's not infallible. I don't have data to quantify the benefit, but it's certainly a lot better than not testing your tests at all. One common failure you'll discover this way is tests which surprise you by passing all the time, even when the code-under-test isn't implemented yet. I know! But somehow it really does keep happening. Also, this will reveal tests which pass even when the code-under-test does the wrong thing. The opposite sort of bad test, one which fails even when it should pass, will be obvious even without TDD.

Who we are ?

The Linux Foundation is a non-profit consortium dedicated to the growth of Linux.

More About the foundation...

Frequent Questions

Join / Linux Training / Board