Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux

pid: generalize task_active_pid_ns

Currently task_active_pid_ns is not safe to call after a task becomes a
zombie and exit_task_namespaces is called, as nsproxy becomes NULL. By
reading the pid namespace from the pid of the task we can trivially solve
this problem at the cost of one extra memory read in what should be the
same cacheline as we read the namespace from.

When moving things around I have made task_active_pid_ns out of line
because keeping it in pid_namespace.h would require adding includes of
pid.h and sched.h that I don't think we want.

This change does make task_active_pid_ns unsafe to call during
copy_process until we attach a pid on the task_struct which seems to be a
reasonable trade off.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Bastian Blank <bastian@waldi.eu.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Nadia Derbey <Nadia.Derbey@bull.net>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

authored by

Eric W. Biederman and committed by
Linus Torvalds
61bce0f1 f9fb860f

+9 -7
+1 -5
include/linux/pid_namespace.h
··· 79 79 } 80 80 #endif /* CONFIG_PID_NS */ 81 81 82 - static inline struct pid_namespace *task_active_pid_ns(struct task_struct *tsk) 83 - { 84 - return tsk->nsproxy->pid_ns; 85 - } 86 - 82 + extern struct pid_namespace *task_active_pid_ns(struct task_struct *tsk); 87 83 void pidhash_init(void); 88 84 void pidmap_init(void); 89 85
+2 -2
kernel/fork.c
··· 1126 1126 1127 1127 if (pid != &init_struct_pid) { 1128 1128 retval = -ENOMEM; 1129 - pid = alloc_pid(task_active_pid_ns(p)); 1129 + pid = alloc_pid(p->nsproxy->pid_ns); 1130 1130 if (!pid) 1131 1131 goto bad_fork_cleanup_io; 1132 1132 1133 1133 if (clone_flags & CLONE_NEWPID) { 1134 - retval = pid_ns_prepare_proc(task_active_pid_ns(p)); 1134 + retval = pid_ns_prepare_proc(p->nsproxy->pid_ns); 1135 1135 if (retval < 0) 1136 1136 goto bad_fork_free_pid; 1137 1137 }
+6
kernel/pid.c
··· 474 474 } 475 475 EXPORT_SYMBOL(task_session_nr_ns); 476 476 477 + struct pid_namespace *task_active_pid_ns(struct task_struct *tsk) 478 + { 479 + return ns_of_pid(task_pid(tsk)); 480 + } 481 + EXPORT_SYMBOL_GPL(task_active_pid_ns); 482 + 477 483 /* 478 484 * Used by proc to find the first pid that is greater than or equal to nr. 479 485 *