[net.micro.68k] Merrily We Optimize

ix222 (04/11/83)

I'd like to patch up a picture of feasible, yet non-obvious
optimizations that a reasonably good 68000 compiler generating
native code might make.  such things as removing invariant
computations from loops and generating no jump instructions
where a branch can be made to work i know about.  can anyone
suggest some useful processor-specific 'tricks'?  Also, can
anyone outline how io is done to the console?

			Much TIA,
			steve serocki
			{ucbvax philabs};sdcsvax;sdccsu3;ix222
			sdcsvax!sdccsu3!ix222 at Ucb-Vax