Arch Linux

'Provides a memory resource controller that manages both anonymous memory and page cache. (See Documentation/cgroups/memory.txt)

Note that setting this option increases fixed memory overhead associated with each page of memory in the system. By this, 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory usage tracking struct at boot. Total amount of this is printed out at boot.

Only enable when you're ok with these trade offs and really sure you need the memory resource controller. Even when you enable this, you can set "cgroup_disable=memory" at your boot option to disable memory resource controller and you can avoid overheads. (and lose benefits of memory resource controller)

This config option also selects MM_OWNER config option, which could in turn add some fork/exit overhead.'

I don't think we want this.

FWIW, memory cgroups were enabled in the 3.5.6 Arch kernel and earlier kernels, possibly since cgroups were introduced. The config options were renamed for 3.6 (see http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commitdiff;h=c255a458055e459f65eb7b7f51dc5dbdd0caf1d8).

I use the MemoryLimit and MemorySoftLimit keys in a couple of systemd service files (see systemd.exec(5)), which rely on the memory cgroups functionality.

Definitively breaks LXC usage under libvirt since the error message states about missing memory controller while starting a LXC container and as Patrick said it was enabled in 3.5 kernels.

Ok will add it again, for 3.6.3.