cgroup: introduce cgroup namespaces
authorAditya Kali <adityakali@google.com>
Fri, 29 Jan 2016 08:54:06 +0000 (02:54 -0600)
committerTejun Heo <tj@kernel.org>
Tue, 16 Feb 2016 18:04:58 +0000 (13:04 -0500)
commita79a908fd2b080977b45bf103184b81c9d11ad07
treefbfb2423937d1ff6ff72e5ca9fa852b0c8d7da98
parent5e2bec7c2248ae27c5b16cd97215ae05c1d39179
cgroup: introduce cgroup namespaces

Introduce the ability to create new cgroup namespace. The newly created
cgroup namespace remembers the cgroup of the process at the point
of creation of the cgroup namespace (referred as cgroupns-root).
The main purpose of cgroup namespace is to virtualize the contents
of /proc/self/cgroup file. Processes inside a cgroup namespace
are only able to see paths relative to their namespace root
(unless they are moved outside of their cgroupns-root, at which point
 they will see a relative path from their cgroupns-root).
For a correctly setup container this enables container-tools
(like libcontainer, lxc, lmctfy, etc.) to create completely virtualized
containers without leaking system level cgroup hierarchy to the task.
This patch only implements the 'unshare' part of the cgroupns.

Signed-off-by: Aditya Kali <adityakali@google.com>
Signed-off-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
fs/proc/namespaces.c
include/linux/cgroup.h
include/linux/nsproxy.h
include/linux/proc_ns.h
kernel/cgroup.c
kernel/cpuset.c
kernel/fork.c
kernel/nsproxy.c