namespaces: Use task_lock and not rcu to protect nsproxy

The synchronous syncrhonize_rcu in switch_task_namespaces makes setns
a sufficiently expensive system call that people have complained.

Upon inspect nsproxy no longer needs rcu protection for remote reads.
remote reads are rare.  So optimize for same process reads and write
by switching using rask_lock instead.

This yields a simpler to understand lock, and a faster setns system call.

In particular this fixes a performance regression observed
by Rafael David Tinoco <rafael.tinoco@canonical.com>.

This is effectively a revert of Pavel Emelyanov's commit
cf7b708c8d1d7a27736771bcf4c457b332b0f818 Make access to task's nsproxy lighter
from 2007.  The race this originialy fixed no longer exists as
do_notify_parent uses task_active_pid_ns(parent) instead of
parent->nsproxy.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>

1 files changed, 6 insertions, 4 deletions

diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 85b62691f4f2..7c6b51a58968 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -373,9 +373,11 @@ struct net *get_net_ns_by_pid(pid_t pid)
         tsk = find_task_by_vpid(pid);
         if (tsk) {
                 struct nsproxy *nsproxy;
-                nsproxy = task_nsproxy(tsk);
+                task_lock(tsk);
+                nsproxy = tsk->nsproxy;
                 if (nsproxy)
                         net = get_net(nsproxy->net_ns);
+                task_unlock(tsk);
         }
         rcu_read_unlock();
         return net;
@@ -632,11 +634,11 @@ static void *netns_get(struct task_struct *task)
         struct net *net = NULL;
         struct nsproxy *nsproxy;
 
-        rcu_read_lock();
-        nsproxy = task_nsproxy(task);
+        task_lock(task);
+        nsproxy = task->nsproxy;
         if (nsproxy)
                 net = get_net(nsproxy->net_ns);
-        rcu_read_unlock();
+        task_unlock(task);
 
         return net;
 }