Opening keynote. Good stuff.
An update on XDP and CPUMap. http://vger.kernel.org/netconf2017_files/XDP_devel_update_NetConf2017_Seoul.pdf
https://lkml.org/lkml/2017/10/10/88 https://lkml.org/lkml/2017/10/10/90 Updated patches: https://lkml.org/lkml/2017/11/4/192 https://lkml.org/lkml/2017/11/4/193
From http://users.ece.utexas.edu/~adnan/pike.html Rule 1. You can’t tell where a program is going to spend its time. Bottlenecks occur in surprising places, so don’t try to second guess and put in a speed hack until you’ve proven that’s where the bottleneck is. Rule 2. Measure. Don’t tune for speed until you’ve measured, and even then don’t […]
https://blog.merovius.de/2017/09/12/diminishing-returns-of-static-typing.html It’s nice to see an argument that doesn’t boil down to ‘more of X is better’. This reminds me of something I heard on the radio many years ago, paraphrasing: A fundamentalist is someone who reacts to contradictory information by saying “we need to X harder/more”
Great post. https://medium.com/@copyconstruct/small-functions-considered-harmful-91035d316c29
https://github.com/qmonnet/tbpoc-bpf Very cool.
https://blog.yadutaf.fr/2017/07/28/tracing-a-packet-journey-using-linux-tracepoints-perf-ebpf/ Very interesting. The key is that eBPF programs can be attached to tracepoints.