@bloodline
I don't think it's so big deal...
from the 68060 errata rev. 4.0:
"The store/load bypass feature, which eliminates a pipeline stall in scenarios of successive instructions storing and loding back data from the same adress, should have very limited performance impact on most programs"
store/load is an optimization made by the pipeline: when you disable it you just add a (some ?) stall (i.e., wait cycle) in the sequence instruction that could be optimized.
My 060 too has this problem, but I don't mind replacing it.
