Tue Jul 26 17:25:19 PDT 2016
P4 3.06ghz from buildworld timing. Also most of the numbers are a lot
lower on the AMD side as far as memory operations and context switches
which seems like a big benefit.
(2) The AMD's syscall overhead is very low compared to the P4.
I ran the 'sc1' test in /usr/src/test/sysperf. The AMD box
comes back at 242nS/loop while the P4 takes 1300nS/loop and
the DELL box (LEAF/SMP) takes 617nS/loop.
getuid() 0.916s 2402600 loops = 0.381uS/loop
on my AMD 1.4ghz.
I think big reason for the speed improvement here is because on every
syscall entry on a P4 you hit a branch misprediction and the AMD it
more about just blowing through the instructions which makes it a lot
at most things.
More information about the Kernel