Skip to content

Commit

Permalink
[PATCH] add child reaper to pid_namespace
Browse files Browse the repository at this point in the history
Add a per pid_namespace child-reaper.  This is needed so processes are reaped
within the same pid space and do not spill over to the parent pid space.  Its
also needed so containers preserve existing semantic that pid == 1 would reap
orphaned children.

This is based on Eric Biederman's patch: http://lkml.org/lkml/2006/2/6/285

Signed-off-by: Sukadev Bhattiprolu <[email protected]>
Signed-off-by: Cedric Le Goater <[email protected]>
Cc: Kirill Korotaev <[email protected]>
Cc: Eric W. Biederman <[email protected]>
Cc: Herbert Poetzl <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
  • Loading branch information
Sukadev Bhattiprolu authored and Linus Torvalds committed Dec 8, 2006
1 parent 6cc1b22 commit 84d7378
Show file tree
Hide file tree
Showing 8 changed files with 40 additions and 19 deletions.
5 changes: 3 additions & 2 deletions fs/exec.c
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,7 @@
#include <linux/binfmts.h>
#include <linux/swap.h>
#include <linux/utsname.h>
#include <linux/pid_namespace.h>
#include <linux/module.h>
#include <linux/namei.h>
#include <linux/proc_fs.h>
Expand Down Expand Up @@ -620,8 +621,8 @@ static int de_thread(struct task_struct *tsk)
* Reparenting needs write_lock on tasklist_lock,
* so it is safe to do it under read_lock.
*/
if (unlikely(tsk->group_leader == child_reaper))
child_reaper = tsk;
if (unlikely(tsk->group_leader == child_reaper(tsk)))
tsk->nsproxy->pid_ns->child_reaper = tsk;

zap_other_threads(tsk);
read_unlock(&tasklist_lock);
Expand Down
5 changes: 3 additions & 2 deletions include/linux/pid.h
Original file line number Diff line number Diff line change
Expand Up @@ -35,8 +35,9 @@ enum pid_type
*
* Holding a reference to struct pid solves both of these problems.
* It is small so holding a reference does not consume a lot of
* resources, and since a new struct pid is allocated when the numeric
* pid value is reused we don't mistakenly refer to new processes.
* resources, and since a new struct pid is allocated when the numeric pid
* value is reused (when pids wrap around) we don't mistakenly refer to new
* processes.
*/

struct pid
Expand Down
6 changes: 6 additions & 0 deletions include/linux/pid_namespace.h
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@ struct pid_namespace {
struct kref kref;
struct pidmap pidmap[PIDMAP_ENTRIES];
int last_pid;
struct task_struct *child_reaper;
};

extern struct pid_namespace init_pid_ns;
Expand All @@ -36,4 +37,9 @@ static inline void put_pid_ns(struct pid_namespace *ns)
kref_put(&ns->kref, free_pid_ns);
}

static inline struct task_struct *child_reaper(struct task_struct *tsk)
{
return tsk->nsproxy->pid_ns->child_reaper;
}

#endif /* _LINUX_PID_NS_H */
1 change: 0 additions & 1 deletion include/linux/sched.h
Original file line number Diff line number Diff line change
Expand Up @@ -1400,7 +1400,6 @@ extern NORET_TYPE void do_group_exit(int);
extern void daemonize(const char *, ...);
extern int allow_signal(int);
extern int disallow_signal(int);
extern struct task_struct *child_reaper;

extern int do_execve(char *, char __user * __user *, char __user * __user *, struct pt_regs *);
extern long do_fork(unsigned long, unsigned long, struct pt_regs *, unsigned long, int __user *, int __user *);
Expand Down
5 changes: 2 additions & 3 deletions init/main.c
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,7 @@
#include <linux/debug_locks.h>
#include <linux/lockdep.h>
#include <linux/utsrelease.h>
#include <linux/pid_namespace.h>
#include <linux/compile.h>

#include <asm/io.h>
Expand Down Expand Up @@ -626,8 +627,6 @@ static int __init initcall_debug_setup(char *str)
}
__setup("initcall_debug", initcall_debug_setup);

struct task_struct *child_reaper = &init_task;

extern initcall_t __initcall_start[], __initcall_end[];

static void __init do_initcalls(void)
Expand Down Expand Up @@ -727,7 +726,7 @@ static int init(void * unused)
* assumptions about where in the task array this
* can be found.
*/
child_reaper = current;
init_pid_ns.child_reaper = current;

cad_pid = task_pid(current);

Expand Down
23 changes: 15 additions & 8 deletions kernel/exit.c
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@
#include <linux/file.h>
#include <linux/binfmts.h>
#include <linux/nsproxy.h>
#include <linux/pid_namespace.h>
#include <linux/ptrace.h>
#include <linux/profile.h>
#include <linux/mount.h>
Expand All @@ -48,7 +49,6 @@
#include <asm/mmu_context.h>

extern void sem_exit (void);
extern struct task_struct *child_reaper;

static void exit_mm(struct task_struct * tsk);

Expand Down Expand Up @@ -260,7 +260,8 @@ static int has_stopped_jobs(int pgrp)
}

/**
* reparent_to_init - Reparent the calling kernel thread to the init task.
* reparent_to_init - Reparent the calling kernel thread to the init task
* of the pid space that the thread belongs to.
*
* If a kernel thread is launched as a result of a system call, or if
* it ever exits, it should generally reparent itself to init so that
Expand All @@ -278,8 +279,8 @@ static void reparent_to_init(void)
ptrace_unlink(current);
/* Reparent to init */
remove_parent(current);
current->parent = child_reaper;
current->real_parent = child_reaper;
current->parent = child_reaper(current);
current->real_parent = child_reaper(current);
add_parent(current);

/* Set the exit signal to SIGCHLD so we signal init on exit */
Expand Down Expand Up @@ -662,7 +663,8 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
* When we die, we re-parent all our children.
* Try to give them to another thread in our thread
* group, and if no such member exists, give it to
* the global child reaper process (ie "init")
* the child reaper process (ie "init") in our pid
* space.
*/
static void
forget_original_parent(struct task_struct *father, struct list_head *to_release)
Expand All @@ -673,7 +675,7 @@ forget_original_parent(struct task_struct *father, struct list_head *to_release)
do {
reaper = next_thread(reaper);
if (reaper == father) {
reaper = child_reaper;
reaper = child_reaper(father);
break;
}
} while (reaper->exit_state);
Expand Down Expand Up @@ -859,8 +861,13 @@ fastcall NORET_TYPE void do_exit(long code)
panic("Aiee, killing interrupt handler!");
if (unlikely(!tsk->pid))
panic("Attempted to kill the idle task!");
if (unlikely(tsk == child_reaper))
panic("Attempted to kill init!");
if (unlikely(tsk == child_reaper(tsk))) {
if (tsk->nsproxy->pid_ns != &init_pid_ns)
tsk->nsproxy->pid_ns->child_reaper = init_pid_ns.child_reaper;
else
panic("Attempted to kill init!");
}


if (unlikely(current->ptrace & PT_TRACE_EXIT)) {
current->ptrace_message = code;
Expand Down
3 changes: 2 additions & 1 deletion kernel/pid.c
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,8 @@ struct pid_namespace init_pid_ns = {
.pidmap = {
[ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL }
},
.last_pid = 0
.last_pid = 0,
.child_reaper = &init_task
};

/*
Expand Down
11 changes: 9 additions & 2 deletions kernel/signal.c
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,9 @@
#include <linux/signal.h>
#include <linux/capability.h>
#include <linux/freezer.h>
#include <linux/pid_namespace.h>
#include <linux/nsproxy.h>

#include <asm/param.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
Expand Down Expand Up @@ -1877,8 +1880,12 @@ int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka,
if (sig_kernel_ignore(signr)) /* Default is nothing. */
continue;

/* Init gets no signals it doesn't want. */
if (current == child_reaper)
/*
* Init of a pid space gets no signals it doesn't want from
* within that pid space. It can of course get signals from
* its parent pid space.
*/
if (current == child_reaper(current))
continue;

if (sig_kernel_stop(signr)) {
Expand Down

0 comments on commit 84d7378

Please sign in to comment.