SOCKMAP – TCP splicing of the future

Re:SOCKMAP – TCP splicing of the future by grandinj on Monday February 18, 2019 @03:11PM (#19193947)

Feels like we're heading to a future where half of userspace will recompile itself as BPF and upload itself into kernel space to be executed there

[ Parent | on HN ]

Re:SOCKMAP – TCP splicing of the future by diegocg on Monday February 18, 2019 @03:48PM (#19194298)

Jokes aside, I wonder how much time will pass until people start writing kernel drivers in BPF

[ Parent | on HN ]

Re:SOCKMAP – TCP splicing of the future by pas on Monday February 18, 2019 @03:52PM (#19194333)

XDP is sort of that for the network subsystem. (Though it doesn't allow you to implement drivers, "just" protocols.) Next step could be filesystem stuff, then libvirt drivers, etc.

But BPF is no that versatile, but at least it has a stable ABI.

[ Parent | on HN ]

Re:SOCKMAP – TCP splicing of the future by agumonkey on Monday February 18, 2019 @04:09PM (#19194483)

let's ask bsd folks who vetted lua

[ Parent | on HN ]

Re:SOCKMAP – TCP splicing of the future by tedunangst on Monday February 18, 2019 @03:59PM (#19194390)

WebAssembly.

[ Parent | on HN ]

Re:SOCKMAP – TCP splicing of the future by nicoburns on Monday February 18, 2019 @09:20PM (#19196188)

I guess you're getting downvotes because you posted this without explanation, but wasm actually makes a lot of sense for safer kernel extensions, as it allows for sandboxing code running as ring 0.

[ Parent | on HN ]

Re:SOCKMAP – TCP splicing of the future by tedunangst on Monday February 18, 2019 @10:17PM (#19196353)

For instance: https://www.wasmjit.org/blog/nginx-on-wasmjit.html

[ Parent | on HN ]