Hi Peter,

I didn't check this much intensively, but I think you are right: it's probably some pipelining bug and the exact type of instruction that changes (or just uses ?) the A register is not so important.
But imagine my situation: I am writing code using Keil C51 compiler and after each run I have to run my home-made utility, which scans binary code and reports 60-100 places where potentiall risk of this bug exists. Of course Keil has nothing to do with it and I've already asked them if they could change instruction sequence generated by their compiler, to avoid this instruction coincidence.

I regret I don't live in USA. I could probably sue DALLLAS for lost of profit then because my project is far behind any schedule because of it ? It is just a sad joke, but anyway, just for my curiosity - could I ?

Best regards,

Mark

P.S. I love this chip! But I want it bug free! NOW!