Guy Harris <g...@alum.mit.edu> wrote:
> On Sep 3, 2014, at 12:34 PM, Michael Richardson <m...@sandelman.ca>
> wrote:
>> It seems that we might need more patches to better select Linux memory
>> mapped packet choices?
> I'd prefer a patch that reduces or the removes the *need* to do so,
> such as this patch, which I'm in the process of testing (yes, I already
> mentioned the problem on linux-netdev, but nobody else has looked much
> at the issues):
My logic is that if we have the switch/env-var to force libpcap to use a
particular method, that permits not only benchmarking, but also isolation of
bugs easier.
--
] Never tell me the odds! | ipv6 mesh networks [
] Michael Richardson, Sandelman Software Works | network architect [
] m...@sandelman.ca http://www.sandelman.ca/ | ruby on rails [
_______________________________________________
tcpdump-workers mailing list
tcpdump-workers@lists.tcpdump.org
https://lists.sandelman.ca/mailman/listinfo/tcpdump-workers
