slub: Do not hold slub_lock when calling sysfs_slab_add()
authorChristoph Lameter <cl@linux.com>
Tue, 17 Jan 2012 15:27:31 +0000 (09:27 -0600)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 2 Apr 2012 16:53:06 +0000 (09:53 -0700)
commit8a9de9cde83d1c913bcbe5bd8c6291f5536a576c
treead0a29e10a50cd8c47daa9aff556ff1afe33b134
parentc45247fdfb2ddf4a96574f8dd3b6857056c20c91
slub: Do not hold slub_lock when calling sysfs_slab_add()

commit 66c4c35c6bc5a1a452b024cf0364635b28fd94e4 upstream.

sysfs_slab_add() calls various sysfs functions that actually may
end up in userspace doing all sorts of things.

Release the slub_lock after adding the kmem_cache structure to the list.
At that point the address of the kmem_cache is not known so we are
guaranteed exlusive access to the following modifications to the
kmem_cache structure.

If the sysfs_slab_add fails then reacquire the slub_lock to
remove the kmem_cache structure from the list.

Reported-by: Sasha Levin <levinsasha928@gmail.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mm/slub.c