* Ulf Magnusson: > Good machine code would be fun to see, though I might need to brush up > on my ARM.
It turns out that ARM doesn't seem to have 8-bit overflow detection, so something like this has to be used (with the arguments in R1 and R2, result in the lower bit of R3): MOV R3, R1, LSL #24 ADDS R3, R2, R3, LSL #24 ADDVS R3, R3, #1 Not sure if this is correct, I don't really know ARM assembly.
