static int
compare_foo (const Gtk::TreeModel::iterator& a, const Gtk::TreeModel::iterator& b)
{
    std::tr1::shared_ptr<Foo> l1, l2;
    l1 = a->get_value (COLUMN_FOO);
    l2 = b->get_value (COLUMN_FOO);

    if (!l1)
        return -1;

    if (!l2)
        return 1;

    return l1->get_id () - l2->get_id ();
}

Can you spot the error? Running it under valgrind indicated that I was reading data from *before* the start of the array that was being sorted. In fact, the following lines of code in g_qsort_with_data() were causing tmp_ptr to be decreased back past the start of the array:

        while ((*compare_func) ((void *) run_ptr, (void *) tmp_ptr, user_data) < 0)
          tmp_ptr -= size;

Hmm. Why is there no guard to prevent tmp_ptr from being decreased past the start? Then I noticed the following comment up a few lines:

    /* Find smallest element in first threshold and place it at the
       array's beginning.  This is the smallest array element,
       and the operation speeds up insertion sort's inner loop. */

So at this point, the code is assuming that the very first element of the array is the smallest, so compare_func should always return >= 0 when it reaches the start of the array. Aha! So the problem is my compare_func. It turns out I forgot to handle the single case of both values being NULL.

    l1 = a->get_value (COLUMN_FOO);
    l2 = b->get_value (COLUMN_FOO);
+
+   if (!l1 && !l2)
+       return 0;

    if (!l1)
        return -1;

So there’s not really a big lesson to be learned here, but if you ever hit strange segfaults deep inside glib’s qsort while sorting a treemodel, do yourself a favor and double-check that your sort function is sane first. Also, do yourself a favor and run it under valgrind as soon as you get a strange segfault. Knowing the exact point of the invalid read is infinitely more helpful than waiting until that invalid read causes a segfault.

deb     http://ppa.launchpad.net/jonathon-jongsma/ubuntu feisty main universe
deb-src http://ppa.launchpad.net/jonathon-jongsma/ubuntu feisty main  universe

Followed by the requisite apt-get install nemiver.

I’ll try to upload new versions periodically. Let me know if you have any issues with them. All of the usual caveats apply — this is development code so may not be completely stable, etc.

By contrast, to get it to work on Joanne’s Windows machine (so she could use it with Skype), it wanted to immediately download updated ‘drivers’ (probably including a bunch of useless utility software — over 100MB!) and of course, install them. It installed without problem, but all-in-all, it took about 10 times as long as it did to get it working on Linux.

When I first started using Linux, I would have never dreamed that hardware support would have gotten this good by now. In fact, on the way home from picking up the webcam, I joked to Joanne that getting it to work on Linux was going to be my weekend project. Ha!

[1] By the way, this is the first time I’ve used Ekiga. It’s quite nice, and the new GUI work looks great.

I’ve been using and developing software on Linux for a little while now, but I’ve always felt slightly uneasy about my lack of detailed knowledge about the Linux kernel. So I picked up rlove’s kernel development book this past weekend and I’m about halfway through it already. I’m gradually beginning to feel more … what’s the opposite of uneasy? easy? at ease? Anyway, it’s quite good and well written. Highly recommended if you’re looking for a nice introduction to the Linux kernel.

Nemiver’s proceeding nicely to a 0.2 release which should be a lot more robust in a lot of ways, including better support for poorly-behaved versions of gdb.

Oh wait, I do have one more complaint about the upgrade: Bug #350053 drives me insane. I love Epiphany, but that bug completely screws up the way I use browse. I can use the ‘smart bookmarks’ feature, but it’s hard to reprogram my brain after getting used to just typing search terms in the location bar and hitting enter. If anyone can get the fix for that into Edgy, I’d be eternally grateful.

Also, if any of the brilliant Tango Friday participants want to make a nice little icon for the Nemiver debugger, that’d be great.

Joanne and I are heading out to the L.A. area for a long weekend tomorrow. Hopefully the weather out there’s nice and warm there. It’s certainly feeling a lot like winter up here. I could really use a few days off work, so I’m hoping for a few nice relaxing days, and the computer’s staying home.

Nevertheless, I’ve been getting a fair number of requests from people who want to use colorscheme but don’t have the necessary expertise to install it from source. And I’d really like to get more people using it and giving me feedback. So I’ve made a couple experimental colorscheme packages for Ubuntu Breezy (i386 and amd64). Hopefully this is just a temporary stop-gap measure until the software gets packaged for more distributions.

I make no guarantee that the packages will even work. Nor will I guarantee any support for them (remember, I don’t actually know what I’m doing — I just whipped those up in an afternoon). So I don’t particularly condone using them, but if you want to use the application and can’t get it installed from source, it’s an option.