Samrobb's Journal: "Today I are a kernel hacker!"

The very first patch I ever submitted to an open source project was in August of 1998. It was a patch to Apache to allow execution of extensionless CGI executables under Windows. It was initially accepted, but then backed out rather quickly because it was horribly broken... but it was accepted, if ever so briefly. It was a really kind of cool.

The overall experience was a good one for me, despite the eventual rejection of the patch. It was obviously enough of an encouragement to keep me going. In the intervening years, I've had submitted patches to a bunch of different projects, with varying degrees of success. Most of them have been relatively minor; but I've tried to (and in most cases, succeeded in) making improvements to things like buildroot, CDT, cygwin, Eclipse, e2fsprogs, and even (once) glibc. Some were minor, some were more significant. All were, effectively, fueled by that first good patch submission experience with the Apache folks.

It's always a nice thing to see a patch accepted, particularly if it means that the problem it addresses will go away in the next release :-) On the other hand, up until this point, I'd never had the chance for a real patch thrill. You know what I mean, right?

Yep - the big kahoona.

The linux kernel .

OK, so the patch I submitted really isn't a kernel patch. It's a patch to the kernel build files.

OK, OK - it's not even a patch to the kernel build files. It's a patch to the kernel configuration utility build files. An itty-bitty little one, at that. Hardly worth noticing, really. Fixes a corner case that only one person in a million will ever encounter.

Except Andrew Morton has picked it up for his -mm tree.

I'm older and wiser than I was in 1998. I like to think that I've learned from my past mistakes. I tested this patch in more ways than one, so I'm fairly sure it won't break things at all, let alone horribly. Still, you can never be sure. Even if it does get backed out for some weird reason... I'll still know that I was good enough to identify the problem, come up with a fix, put together a patch and explain that patch well enough to get it into the pipeline for someone to look at.

Yah, I realize I'm not a real kernel hacker. I work with guys who are, and I know I'm not even in the ballpark on this one compared to some of the things they deal with every day. But I'm a step closer than I was yesterday, and who knows what I'll be doing in a couple of years?

Maybe by then, I really will be able to say "I am a kernel hacker." Now that would be kind of cool.