Some of these things have been around for many years. At least those things that I have heard of, no idea about the ones I have not heard of. Btrfs for example has been in the Linux kernel for almost a decade, and started in Oracle.
Facebook sets Linux kernel tools free
After years of making the world more open and connected – to everyone's delight – Facebook recently moved on to bringing the world closer together. Amid its pursuit of global compression, the data harvesting biz nonetheless developed a handful of difficult-to-pronounce Linux kernel components to make the open source operating …
COMMENTS
-
-
Thursday 1st November 2018 12:22 GMT Anonymous Coward
Re: oomd - for reigning in developers who shouldn't be
The OOM kills processes at random.
OOMD will let you kill a specific process, possibly allowing you to fix the problem.
FB after all is famously a user of "saturate, log, and leak" so choosing which process to restart under such circumstances would be most useful.
-
-
Thursday 1st November 2018 14:06 GMT Doctor Syntax
Who's been pottering about with kernel code?
"The software allows bytecode to run in response to specific events for the purpose of modifying and extending kernel behavior."
Just what everyone needs - send bytecodes to
yoursomebody's kernel and get them executed with, presumably, kernel privileges.-
-
Thursday 1st November 2018 16:04 GMT Anonymous Coward
Re: Who's been pottering about with kernel code?
Yeah, no.
man bpf(2) will make this clearer.
BPF is what libpcap understands and exposes as tcpdump/wireshare filter syntax.
eg. 'dst port 80'
eBPF allows you to have a stat collection buffer in kernel space, and then collect the stats from userspace.
This is huge for lots of tedious bit twiddly reasons, see https://blog.cloudflare.com/epbf_sockets_hop_distance/ for someone with a better explanation.
-
-
Thursday 1st November 2018 15:58 GMT Anonymous Coward
Re: Who's been pottering about with kernel code?
Its not as bad as all that.
EBPF is a verified safe subset of instructions that can be attached to certain kernel paths.
Think a very fine grained ptrace with much lower overhead...
As this is a rather esoteric subject, I can recommend a very nice blog post series from
Cloudflare on the subject.
This is by far the most accessible of the series https://blog.cloudflare.com/epbf_sockets_hop_distance/
-
-