> The L4 HotOS 97 paper "Achieved IPC Performance" claims <1uS IPCs on > Pentium-166s. Is this bogus? If not, why is EROS fast-path IPC pipe > latency an order of magnitude slower at 18uS? (I know it involves > several IPCs, but ten?) One other bit on that. The EROS pipe test is using cruddy stub code that rougly doubles the IPC times. I think this reinforces the message of that paper.