Testing interrupted

Okay, I’m giving up.

I know that I really should complete this test suite, but I just can’t at the moment.

It’s been a long time since I started it, and I’ve been progressing at a snail pace. No, not exactly : I’ve made a lot of progresses first, but now nothing new’s happening. I’ve corrected all major, easy-to-spot flaws of the code, now all that’s left to do is to boringly implement all test cases, so that memory management modifications in the future can be quickly checked for correctness. It’s important, but hardly exciting.

The problem is that now, in order to preserve my mental sanity and motivation, I would really like to breathe a bit of fresh air. I just have to postpone this at the moment and work on something more interesting. That something will be interrupt management, and maybe a few tricks based on it after that (clock and keyboard driver, maybe early scheduling work…). Once I’ve got some exciting results, I’ll sure have to go back to work on enforcing correctness before creating a new hackish open-source kernel.

Sure, that’s a dangerous track to hack now and polish later, and I should be more careful. But I just want to have fun at the moment :) Want some holidays in my coding process. Is it such a bad thing ?

EDIT : Oh, and of course the doc has been updated to reflect changes to the memory management system. After all, I should break the API as much as necessary while I still can ^^

3 thoughts on “Testing interrupted

  1. Michael Gorman February 27, 2011 / 3:49 pm

    Dear Mr Grasland,
    I reply to your article in OS News, Interrupt Mechanisms on non-x86 Architectures. I reply here because, although I am a long time reader of OSN, it screwed up on my membership application after it introduced compulsory membership. Therefore I cannot post this reply there.

    My suggestion is that you consider the interrupt architecture on the PDP-11 and VAX series computers, partly because it is (or certainly was) well documented, and partly because it was a neat architecture.

    My own memory of these things is probably unreliable – I retired some years ago – and so I cop out by referring you to “documentation”. I wish you good luck in your project and look forward to reading more of it on OSN.

  2. Hadrien February 27, 2011 / 4:04 pm

    You probably will hear more about this project, when more of it is done :)

    At the moment, if I was to write an article about this project, it would be to say “Hello, here I am, and here’s what I want to achieve, and how I plan to achieve it”. There’s not enough implemented code for it to be worth an article yet.

    I’ve tried to write such articles some times, but with no satisfactory result so far. So at the moment, I wait until I have implemented more interesting code.

    Anyway, I’ll have a look at the computers you mentioned. Indeed, documentation seems to be abundant so far.

  3. Jérôme DUMESNIL March 2, 2011 / 11:52 am

    Hi Hadrien,

    I completely agree with you :
    – Testing is absolutely necessary
    – Testing is boring ;-)
    – Implementing memory management, interrupt handling, … is from far away a lot more interesting than testing !!!

    Have fun :-) !!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s