Muli Ben-Yehuda's journal

January 6, 2004

Filed under: Uncategorized — Muli Ben-Yehuda @ 9:59 PM

This entry will be disjointed. You have been warned. I stare at this empty xemacs buffer and an insane urge to fill it with words fills me. So many words that they spill over the brim. A brain dump, if you will. ./braindup -l (work joke, not funny, ha ha). Go on reading at your own peril.

The hard disk on my faithful work laptop hydra died. A new hard disk was requested and received. The system team installed a customized Fedora on it (based on FC1, I think). So far, I like it. With caveats. Old habits die hard. I’m still customizing things and setting them up the way I like them, where possible, or finding new things to like, where impossible. gnome with metacity won’t let me drag windows to other desktops! *scoff*.

Projects and funding for next year are still very much in flux at work and nothing is decided. Things should be converging towards the end of the month. Keep your fingers crossed for my research proposal getting funded. Then I will work on Cool Kernel Stuff(TM). Alternatively, keep your fingers crossed for me migrating over to The Other Project. Then I will work on Cool Kernel Stuff(TM as well. Anything (well, not really) will be better than the current mildly-interesting-but-no-kernel-hacking-required project.

A guy I knew in the army, N.A., interviewed here today. We spent some time talking. It’s funny, the more things change, the more they stay the same. He was wearing a lovely suit. I wish I wore suits to work.

Amir is hacking on syscalltrack’s FreeBSD port, and I want to help him with the Linux bits, but not finding the time. There’s also the shpte code that’s still waiting for a de-bugging. Maybe tonight? maybe. After I finish testing trident in -mm1-rc1. Maybe also give -tiny a spin? there’s an old 486 comatose under the second desk that would love a tiny penguin invasion.

The other day, in a bind to fix a bug quickly in someone else’s code that was building a char* array and passing that to vprintf and friends as a va_list, I wrote this gem:

 static void my_vsprintf(char* dest, const char* fmt, char** pva) 
        size_t num_fmts = count_fmts(fmt);

        switch(num_fmts) { 
                case 0: sprintf(dest, fmt); break; 
                case 1: sprintf(dest, fmt, pva[0]); break; 
                case 2: sprintf(dest, fmt, pva[0], pva[1]); break; 
                case 3: sprintf(dest, fmt, pva[0], pva[1], pva[2], pva[3]); break; 
                /* more of the same */ 
                default: assert(0 * num_fmts); 

You may barf now. I think that was a low point of my professional programming career. I wonder what the fuck where they thinking, build va_lists by hand.

Tomorrow is the last day of the Linux sysprogramming course. We are going to learn to compile kernels. Insert snarky comment here. That should be quality hacking time… during the last class I waged battle against the combined forces of Oleg’s bugzilla and apache 2.0, and won. What will I fight tomorrow? I could take on shpte, but that’s a tall order. Nonetheless, I could use the ensuing humility.

Time to go home, me’thinks.


  1. The opinion of most folks I’ve spoken to is that Gnome 1 was a major improvement on Gnome 2. (I personally switched to WindowMaker when G2 became “standard”. Feh.)

    Comment by geekosaur — January 6, 2004 @ 1:21 PM | Reply

    • There’s a huge on-going thread on linux-elitists about GNOME. Personally, as long as it gets the work done, I’ll use it, but it’s edging closer and closer to the point where it’s not getting the work done, replacing functionality with eye candy.

      Comment by mulix — January 6, 2004 @ 2:31 PM | Reply

      • That, in a nutshell, is the problem and the reason I dropped it. So far evolution is the only GNOME app that doesn’t seem to have gotten stupider in later versions… and I’m not sure how long that will last. 😦

        Comment by geekosaur — January 6, 2004 @ 2:34 PM

  2. having problems with shpte code.

    Comment by ideawerkz — January 6, 2004 @ 8:25 PM | Reply

    • you and me both. I’m starting to think of rewriting it from scratch, after hearing wli’s scathing comments and considering the instructive value of both options.

      Comment by mulix — January 7, 2004 @ 12:02 AM | Reply

      • I was there when I read wli’s scathing feedbacks on the code. And likewise, I think it might be a better approach to rewrite while keeping in mind what was done in the original patch. What do you have in mind?

        Comment by ideawerkz — January 7, 2004 @ 1:42 AM

      • Nothing particularly clever for the time being. I should probably invest some time in a design before starting to code, but I also like jumping in and learning as I go along, and then going back and rewriting things after the gained experience. Do you have anything specific in mind?

        Comment by mulix — January 7, 2004 @ 1:46 AM

RSS feed for comments on this post. TrackBack URI

Leave a Reply

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

You are commenting using your 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 )

Connecting to %s

Blog at

%d bloggers like this: