Optimized multibyte multiplication code

Hi guys,

I implemented some optimized 16bit x 16bit multiplications, 32x16 and also 16x8constant, it's pretty useful if you need to perform some digital signal processing on Arduino. Here's the code http://www.math.ucla.edu/~npozar/arduino/mult.zip and some description http://mekonik.wordpress.com/2009/03/18/arduino-avr-gcc-multiplication/

I'm gonna expand it if there's interest. Hope somebody finds it useful.

Best, Norbert

Assembler is too foreign for me, but I'm not scared of learning a new (computer!) language. What would be a good place to look for documentation/tutorial for the cryptic GNU inline assembler syntax? The MCU datasheet knows all about the instructions, but I'm having a hard time putting them together, and I hate it when I don't fully understand something ...