Title says it all.
Are there processors with FPUs that deal with denorms arithmetically correctly (don't assume they're zero nor flush them to zero) and don't cause an exception? I would be interested in knowing who the black sheep and the white sheep are. Like the Intel Xeon or AMD Opteron, are denorms for them just any old 64-bit floating-point number?
My app interest is about real-time digital signal processing.