bla repository

In the meantime I wrote kind of useless software but haven't got time before to publish it. A program to send modbus (ASCII or RTU) commands or terminated text strings over UDP or serial connection.

I've installed pam-ppp, tested, corrected, liked it and recently - rewritten totally from scratch to implement more features and to correct some pitfalls. I called this program "OTPasswd" and I regard it's cool. Why?

Being forced from sometime to login into my system with SSH from a machine I wouldn't ever trust (winxp/vista without firewall, publicly accessible) motivated me to configure an One-time Password system.

UPDATE: I've written similar system completely from scratch using ideas from ppp-pam. It is called "OTPasswd", since the name doesn't seem to be taken and its pretty neat shortcut for One-Time Passwords. It's actually pretty advanced, version in repository supports OOB messaging, policies, has nice manuals, etc. See entry on otpasswd. And homepage at Savannah: OTPasswd.

Name? bla.
Sex? Male.
Hobby? Raytracing!
Duh. ;-/

Another program realized for my university; this time the main idea was to write something multitasking in assembler of x86. The program raytraces only quaternion julia fractals and is much more fun from benchmarking point of view. Program was written in C first, then using OProfile I've checked which functions take the most of the CPU time and those got rewritten in asm.

Algorithm has C version (x86/x64), optimized SSE2 assembler (x86), plain CUDA version (tested on x64 only, no crosscompilation) and mixed version - CUDA + C. They are selectable using compile-time macros. What makes it more fun is that the C version can be compiled by GCC using either x87 arithmetics or (naive) SSE2. Pictures, links and benchmark summary follow.

I've just spent a weekend forcing RFM12 modules to do what I need. RFM12 are small digital radio transceivers built using RF12 chips. Their obvious advantages are price, quite nice range (100m with a 433Mhz version working with low bitrate) and no external components at all (Ok: except for one, tiny, negligible resistor). In the mode I used them they are just a cheap SPI->Antenna solution. However this text would end here if not for their most important disadvantages: inaccurate documentation with insane code examples, lack of any higher level layer and half-duplex operation.

Further explanation and code follow.

It was a very long-timed project, albeit not a very intensive one. The robot was developed pretty slowly: if I'd place photos of each version in a line it would resemble homo sapiens evolution chart... the first and the current version differ even more then monkey and human. One film, some source code and brief description follow.

I finally decided to give tilling, keyboard-controlled window managers a try. First I've tried few (most notably WMII, Ion and xmonad), then chosen one and been configuring it few hours (a one day)... That's how I stepped on the path with no return, because, seriously --- when you've created good keybindings, learned them and can use them fluently is there any reason to stick with classical window managers? Especially on laptops, when there's not always enough time to plug a mouse...

I've switched from PSI to Pidgin. Mostly because there were no Gentoo extras for 0.11, still no scripting support and no plugin support at all. Nevertheless vanilla Pidgin had the same issue as vanilla PSI - no status logging. Also I couldn't find any plugin which would do the work. Maybe I've missed something... but why don't people want to be able to see someone statuses from the past week? Why is it so hard to make such a simple functionality?

I've finished a work on my Computer Programming assignment for studies few months ago but haven't had time to publish it... so without further introduction all related files are here.

During my current robotic project I was using ATTiny26 with L298 motor controller on one universal board, accelerometer on another board (home etched), serial level converter (max232) on yet another universal board and for programming I was using another (fourth) board...

It was modular - ok, but I got tired with all the wires, problems, keeping code below 2KB level and that's why I designed, ordered, soldered and tested new board implementing all required features. I just wanted to share the PCB schemas and ideas. Keep in mind that it's mine third board design ever. Previous designs shall not be commented (I know they are awful)...

In previous post I already mentioned that I'm going to write in incoming semester a simple ray tracer in C++ as a "programming project" (already finished, description is here). As C++ is not mine favorite language out there I've decided to give ray tracing a try with Objective Caml. Brief illustrated history follows...

There's nothing that makes me more happy than some new embedded device which can run Linux...

I've started working with Linux port on HP iPAQ hx2790 device. All already existing images for hx2000 family had issues with screen (image was right shifted). I've emerged cross toolchain and installed OpenEmbedded tree. After some mangling the screen issue was solved and kernel prepared for future modifications. It would be cool to run IrDA, BT and Wi-fi... for now I'm fighting with building images with OE.

Preliminary images can be found here

There are many networks where connection is limited to HTTP, FTP over Proxies (Academical networks; publicly available wi-fi's etc.) or where http proxy has much faster connection than other protocols.

I also wanted to review Linux networking and write some general-usage tool. And so I did...

Two (almost three) months of time for anything. Two AVR projects waiting to be finished, server migration still in progress, C++ project for studies (raytracer) just started, OCaml documentation still not finished...

I'm good at making long-term plans but rather poor in their finalization. Just after finishing some of the mentioned things I'm going to publish some info.

What can you do with a free weekend, yacc+lex compilers and a stupid idea? Just another highly requested esoteric language I suppose.

This one is based on a slang of one of the Polish street subcultures (the one with a special love to the track suits. Track suit = "Dres"). It employs a rather dirty language. Language specification and the idea itself was developed about a year ago by me and my two colleagues (tmd and shm) on IRC, but we were lacking the necessary wisdom and a will to implement it.

Language is called "Dress". Its compiler is able to compile conditional expressions (branching and loops), few register operations (inc/dec/set) and simple I/O into x86 assembler. It's by no means complete but I doubt if I'll do something more with this compiler. I just wanted to prove it's not so difficult to implement.

Theoretically it should be easy to port language specification so it would be based on a facultative dialect.

I'd like to proudly announce availability of the fresh release of NVM (I'm known of overcomplicated sentences, especially in English. Pity, that this overcomplexity doesn't come with grammatical correctness).

NVM stands for NVM Virtual Machine (yes, right. It's the still-and-forever popular recursive acronym). It's a small, quite portable and as I hope fast virtual machine. I've written it without precise purpose as a work for studies but I'm considering using it for scripting one of my µC projects (mobile robot).

Indeed: why? Why have I decided to start site like this?