Not to forget per-process fs namespace, and text-only replacement of ugly ioctls.
Re:Plan 9 ported to the Raspberry Pi
by justincormack on Saturday August 10, 2013 @06:11AM
(#6190585)
Linux does have per process fs namespaces now... ioctls on the other hand...
In Linux CLONE_NEWNS requires CAP_SYS_ADMIN which makes it almost completely useless; it can't be used to the extent and in the way it is used in Plan 9.
I believe once you create a user namespace (https://lwn.net/Articles/528078/) you can then create the other kinds of namespaces (mount, process, etc.) without requiring privileges outside of your user namespace.
LWN ran a great series on namespaces recently: part 5 (https://lwn.net/Articles/532593/) covers user namespaces.
Re:Plan 9 ported to the Raspberry Pi
by justincormack on Saturday August 10, 2013 @08:58AM
(#6190933)
I think user namespaces allow that without root. Haven't tried yet though...
Mostly I think about union mounts (which are very different in plan9 than the version that's been integrated into Linux) and the privilege escalation system (which has always seemed, to me, to be so much better than the sieve that is setuid).
(Note: I'm aware there have been attempts to bring purer versions of these things to linux, but they never gain any traction)