Hacker News Reader

Something funny about the numbers here. The assembly stub took 8ms, but the whole program only takes 1ms?

That was a typo -- the assembly stub took about 0.8 ms as shown in the trace.

Thanks, fixed.