Pete's Log: pim thoughts
Entry #1131, (Coding, Hacking, & CS stuff)(posted when I was 23 years old.)
programs do not randomly segfault. there must be a race condition in pilsner somewhere. It segfaulted on me under unusual circumstances today, and try as I might, I can't reproduce the segfault. But since the unusual circumstances under which the segfault occured were circumstances I didn't want happening anyway, I simply modified the code to not allow those circumstances.
I've resigned myself to the fact that PIM Lite is going to be slower than an intoxicated walrus. That's just the way it's gonna have to be, and that's ok, since it's a prototype system anyway. So I'll plug away at slow prototype OS code, but I will also focus on the future and attempt to get involved in the ISA design of the real PIM chip in order to ensure it doesn't find itself as crippled as PIM Lite is. The number of system calls that PIM Lite programs will have to make in order to accomplish anything reasonable is going to be absurd.
I'm becoming frustrated with the lack of communication that seems to be going on in our department. My biggest gripe right now is that PIM seems to be a lost soul, wandering through the post-apocalyptic world that the apparent demise of HTMT has left us in. PIM knows it has potential, but without a guiding force to steer it towards a specific goal, PIM is having a tough time figuring out what it wants to be. This irritates me. It's hard to weigh tradeoffs when you have no ultimate motive. But alas, I still enjoy it. Cuz I've got PIM fever!