aboutsummaryrefslogtreecommitdiff
path: root/kernel/nsproxy.c
diff options
context:
space:
mode:
authorEric W. Biederman <ebiederm@xmission.com>2010-03-02 14:51:53 -0800
committerEric W. Biederman <ebiederm@xmission.com>2012-11-19 05:59:09 -0800
commit17cf22c33e1f1b5e435469c84e43872579497653 (patch)
treec5cec307c06f7e4ddc26e04e73c8833897b67113 /kernel/nsproxy.c
parent49f4d8b93ccf9454284b6f524b96c66d8d7fbccc (diff)
pidns: Use task_active_pid_ns where appropriate
The expressions tsk->nsproxy->pid_ns and task_active_pid_ns aka ns_of_pid(task_pid(tsk)) should have the same number of cache line misses with the practical difference that ns_of_pid(task_pid(tsk)) is released later in a processes life. Furthermore by using task_active_pid_ns it becomes trivial to write an unshare implementation for the the pid namespace. So I have used task_active_pid_ns everywhere I can. In fork since the pid has not yet been attached to the process I use ns_of_pid, to achieve the same effect. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Diffstat (limited to 'kernel/nsproxy.c')
-rw-r--r--kernel/nsproxy.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c
index ca27d2c5264..acc92680381 100644
--- a/kernel/nsproxy.c
+++ b/kernel/nsproxy.c
@@ -84,7 +84,7 @@ static struct nsproxy *create_new_namespaces(unsigned long flags,
goto out_ipc;
}
- new_nsp->pid_ns = copy_pid_ns(flags, task_cred_xxx(tsk, user_ns), task_active_pid_ns(tsk));
+ new_nsp->pid_ns = copy_pid_ns(flags, task_cred_xxx(tsk, user_ns), tsk->nsproxy->pid_ns);
if (IS_ERR(new_nsp->pid_ns)) {
err = PTR_ERR(new_nsp->pid_ns);
goto out_pid;