summaryrefslogtreecommitdiffstats
path: root/fs/nfs
diff options
context:
space:
mode:
authorKirill A. Shutemov <kirill.shutemov@linux.intel.com>2012-09-26 03:33:07 +0200
committerAl Viro <viro@zeniv.linux.org.uk>2012-10-03 03:35:55 +0200
commit8c0a85377048b64c880e76ec7368904fe46d0b94 (patch)
tree12e075f4159e88b09a89e35866017f82f80bcc7d /fs/nfs
parentbtrfs: reada_extent doesn't need kref for refcount (diff)
downloadlinux-8c0a85377048b64c880e76ec7368904fe46d0b94.tar.xz
linux-8c0a85377048b64c880e76ec7368904fe46d0b94.zip
fs: push rcu_barrier() from deactivate_locked_super() to filesystems
There's no reason to call rcu_barrier() on every deactivate_locked_super(). We only need to make sure that all delayed rcu free inodes are flushed before we destroy related cache. Removing rcu_barrier() from deactivate_locked_super() affects some fast paths. E.g. on my machine exit_group() of a last process in IPC namespace takes 0.07538s. rcu_barrier() takes 0.05188s of that time. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Diffstat (limited to 'fs/nfs')
-rw-r--r--fs/nfs/inode.c5
1 files changed, 5 insertions, 0 deletions
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 9b47610338f5..e4c716d374a8 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1571,6 +1571,11 @@ static int __init nfs_init_inodecache(void)
static void nfs_destroy_inodecache(void)
{
+ /*
+ * Make sure all delayed rcu free inodes are flushed before we
+ * destroy cache.
+ */
+ rcu_barrier();
kmem_cache_destroy(nfs_inode_cachep);
}