HEADS UP: zerocopy bpf commits impending
rwatson at FreeBSD.org
Wed Apr 23 10:05:26 UTC 2008
On Wed, 23 Apr 2008, Darren Reed wrote:
> Out of curiosity, were those numbers for single cpu/core systems or systems
> with more than one cpu/core active/available?
> I know the testing I did was all single threaded, so moving time from kernel
> to user couldn't be expected to make a large overall difference in a non-SMP
> kernel (NetBSD-something at the time.)
I believe all multi-core.
BTW, if you are set up to do performance measurement on BPF, we'd really love
to see further feedback relating to successful or unsuccessful measurement.
Robert N M Watson
University of Cambridge
More information about the freebsd-arch