pid namespaces: define is_global_init() and is_container_init()

is_init() is an ambiguous name for the pid==1 check.  Split it into
is_global_init() and is_container_init().

A cgroup init has it's tsk->pid == 1.

A global init also has it's tsk->pid == 1 and it's active pid namespace
is the init_pid_ns.  But rather than check the active pid namespace,
compare the task structure with 'init_pid_ns.child_reaper', which is
initialized during boot to the /sbin/init process and never changes.

Changelog:

	2.6.22-rc4-mm2-pidns1:
	- Use 'init_pid_ns.child_reaper' to determine if a given task is the
	  global init (/sbin/init) process. This would improve performance
	  and remove dependence on the task_pid().

	2.6.21-mm2-pidns2:

	- [Sukadev Bhattiprolu] Changed is_container_init() calls in {powerpc,
	  ppc,avr32}/traps.c for the _exception() call to is_global_init().
	  This way, we kill only the cgroup if the cgroup's init has a
	  bug rather than force a kernel panic.

[akpm@linux-foundation.org: fix comment]
[sukadev@us.ibm.com: Use is_global_init() in arch/m32r/mm/fault.c]
[bunk@stusta.de: kernel/pid.c: remove unused exports]
[sukadev@us.ibm.com: Fix capability.c to work with threaded init]
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Acked-by: Pavel Emelianov <xemul@openvz.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Herbert Poetzel <herbert@13thfloor.at>
Cc: Kirill Korotaev <dev@sw.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

32 files changed, 52 insertions, 37 deletions

diff --git a/arch/alpha/mm/fault.c b/arch/alpha/mm/fault.c
index 25154df3055a..e0593e606140 100644
--- a/arch/alpha/mm/fault.c
+++ b/arch/alpha/mm/fault.c
@@ -188,7 +188,7 @@ do_page_fault(unsigned long address, unsigned long mmcsr,
         /* We ran out of memory, or some other thing happened to us that
            made us unable to handle the page fault gracefully.  */
  out_of_memory:
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index 59ed1d05b71b..a8a7dab757eb 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -197,7 +197,7 @@ survive:
         return fault;
 
 out_of_memory:
-        if (!is_init(tsk))
+        if (!is_global_init(tsk))
                 goto out;
 
         /*
diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c
index 9a73ce7eb50f..8a7caf8e7b45 100644
--- a/arch/avr32/kernel/traps.c
+++ b/arch/avr32/kernel/traps.c
@@ -89,7 +89,7 @@ void _exception(long signr, struct pt_regs *regs, int code,
          * generate the same exception over and over again and we get
          * nowhere.  Better to kill it and let the kernel panic.
          */
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 __sighandler_t handler;
 
                 spin_lock_irq(&current->sighand->siglock);
diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c
index 11472f8701bd..6560cb18b4e3 100644
--- a/arch/avr32/mm/fault.c
+++ b/arch/avr32/mm/fault.c
@@ -160,7 +160,7 @@ bad_area:
                 if (exception_trace && printk_ratelimit())
                         printk("%s%s[%d]: segfault at %08lx pc %08lx "
                                "sp %08lx ecr %lu\n",
-                               is_init(tsk) ? KERN_EMERG : KERN_INFO,
+                               is_global_init(tsk) ? KERN_EMERG : KERN_INFO,
                                tsk->comm, tsk->pid, address, regs->pc,
                                regs->sp, ecr);
                 _exception(SIGSEGV, regs, code, address);
@@ -209,7 +209,7 @@ no_context:
          */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
@@ -231,7 +231,7 @@ do_sigbus:
         if (exception_trace)
                 printk("%s%s[%d]: bus error at %08lx pc %08lx "
                        "sp %08lx ecr %lu\n",
-                       is_init(tsk) ? KERN_EMERG : KERN_INFO,
+                       is_global_init(tsk) ? KERN_EMERG : KERN_INFO,
                        tsk->comm, tsk->pid, address, regs->pc,
                        regs->sp, ecr);
 
diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c
index 32f26253c4e8..7571076a16a1 100644
--- a/arch/ia64/mm/fault.c
+++ b/arch/ia64/mm/fault.c
@@ -274,7 +274,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re
 
   out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/m32r/mm/fault.c b/arch/m32r/mm/fault.c
index 70a766aad3e0..4a71df4c1b30 100644
--- a/arch/m32r/mm/fault.c
+++ b/arch/m32r/mm/fault.c
@@ -271,7 +271,7 @@ no_context:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(tsk)) {
+        if (is_global_init(tsk)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/m68k/mm/fault.c b/arch/m68k/mm/fault.c
index eaa618681159..f493f03231d5 100644
--- a/arch/m68k/mm/fault.c
+++ b/arch/m68k/mm/fault.c
@@ -180,7 +180,7 @@ good_area:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c
index 5699c7713e2f..fa636fc6b7b9 100644
--- a/arch/mips/mm/fault.c
+++ b/arch/mips/mm/fault.c
@@ -173,7 +173,7 @@ no_context:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(tsk)) {
+        if (is_global_init(tsk)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index bf9e39c6e296..9fb4a6849c5a 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -201,7 +201,7 @@ void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr)
          * generate the same exception over and over again and we get
          * nowhere.  Better to kill it and let the kernel panic.
          */
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 __sighandler_t handler;
 
                 spin_lock_irq(&current->sighand->siglock);
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index ab3546c5ac3a..a18fda361cc0 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -375,7 +375,7 @@ bad_area_nosemaphore:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c
index 3a393c7f390e..a1ab25c7082f 100644
--- a/arch/powerpc/platforms/pseries/ras.c
+++ b/arch/powerpc/platforms/pseries/ras.c
@@ -332,7 +332,7 @@ static int recover_mce(struct pt_regs *regs, struct rtas_error_log * err)
                    err->disposition == RTAS_DISP_NOT_RECOVERED &&
                    err->target == RTAS_TARGET_MEMORY &&
                    err->type == RTAS_TYPE_ECC_UNCORR &&
-                   !(current->pid == 0 || is_init(current))) {
+                   !(current->pid == 0 || is_global_init(current))) {
                 /* Kill off a user process with an ECC error */
                 printk(KERN_ERR "MCE: uncorrectable ecc error for pid %d\n",
                        current->pid);
diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c
index 3f3b292eb773..c78568905c3b 100644
--- a/arch/ppc/kernel/traps.c
+++ b/arch/ppc/kernel/traps.c
@@ -121,7 +121,7 @@ void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr)
          * generate the same exception over and over again and we get
          * nowhere.  Better to kill it and let the kernel panic.
          */
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 __sighandler_t handler;
 
                 spin_lock_irq(&current->sighand->siglock);
diff --git a/arch/ppc/mm/fault.c b/arch/ppc/mm/fault.c
index 94913ddcf76e..254c23b755e6 100644
--- a/arch/ppc/mm/fault.c
+++ b/arch/ppc/mm/fault.c
@@ -290,7 +290,7 @@ bad_area:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/s390/lib/uaccess_pt.c b/arch/s390/lib/uaccess_pt.c
index 60604b2819b2..b159a9d65680 100644
--- a/arch/s390/lib/uaccess_pt.c
+++ b/arch/s390/lib/uaccess_pt.c
@@ -64,7 +64,7 @@ out:
 
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 14c241ccdd4d..2456b52ed068 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -211,7 +211,7 @@ static int do_out_of_memory(struct pt_regs *regs, unsigned long error_code,
         struct mm_struct *mm = tsk->mm;
 
         up_read(&mm->mmap_sem);
-        if (is_init(tsk)) {
+        if (is_global_init(tsk)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 return 1;
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 4729668ce5bf..f33cedb353fc 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -207,7 +207,7 @@ no_context:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/sh64/mm/fault.c b/arch/sh64/mm/fault.c
index dd81c669c79b..7aea586fc3d0 100644
--- a/arch/sh64/mm/fault.c
+++ b/arch/sh64/mm/fault.c
@@ -278,7 +278,7 @@ bad_area:
                         show_regs(regs);
 #endif
                 }
-                if (is_init(tsk)) {
+                if (is_global_init(tsk)) {
                         panic("INIT had user mode bad_area\n");
                 }
                 tsk->thread.address = address;
@@ -320,14 +320,14 @@ no_context:
  * us unable to handle the page fault gracefully.
  */
 out_of_memory:
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 panic("INIT out of memory\n");
                 yield();
                 goto survive;
         }
         printk("fault:Out of memory\n");
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index bd060551e619..cb3321f8e0a9 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -108,7 +108,7 @@ out_nosemaphore:
  * us unable to handle the page fault gracefully.
  */
 out_of_memory:
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 up_read(&mm->mmap_sem);
                 yield();
                 down_read(&mm->mmap_sem);
diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c
index 9f38b12b4af1..8bab2b2efaff 100644
--- a/arch/x86/lib/usercopy_32.c
+++ b/arch/x86/lib/usercopy_32.c
@@ -748,7 +748,7 @@ survive:
                         retval = get_user_pages(current, current->mm,
                                         (unsigned long )to, 1, 1, 0, &pg, NULL);
 
-                        if (retval == -ENOMEM && is_init(current)) {
+                        if (retval == -ENOMEM && is_global_init(current)) {
                                 up_read(&current->mm->mmap_sem);
                                 congestion_wait(WRITE, HZ/50);
                                 goto survive;
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index 6555c3d14371..4fc5e400cf0a 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -587,7 +587,7 @@ no_context:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(tsk)) {
+        if (is_global_init(tsk)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index 5e0e54906c48..5149ac136a5d 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -554,7 +554,7 @@ no_context:
  */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 goto again;
         }
diff --git a/arch/xtensa/mm/fault.c b/arch/xtensa/mm/fault.c
index 2f842859948f..33f366be323f 100644
--- a/arch/xtensa/mm/fault.c
+++ b/arch/xtensa/mm/fault.c
@@ -145,7 +145,7 @@ bad_area:
          */
 out_of_memory:
         up_read(&mm->mmap_sem);
-        if (is_init(current)) {
+        if (is_global_init(current)) {
                 yield();
                 down_read(&mm->mmap_sem);
                 goto survive;
diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index 78d14935f2b8..de60e1ea4fb3 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -251,7 +251,7 @@ static void send_sig_all(int sig)
         struct task_struct *p;
 
         for_each_process(p) {
-                if (p->mm && !is_init(p))
+                if (p->mm && !is_global_init(p))
                         /* Not swapper, init nor kernel thread */
                         force_sig(sig, p);
         }
diff --git a/include/linux/sched.h b/include/linux/sched.h
index df6049e5e8bf..47cf81d62047 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1237,12 +1237,20 @@ static inline int pid_alive(struct task_struct *p)
 }
 
 /**
- * is_init - check if a task structure is init
+ * is_global_init - check if a task structure is init
  * @tsk: Task structure to be checked.
  *
  * Check if a task structure is the first user space task the kernel created.
+ *
+ * TODO: We should inline this function after some cleanups in pid_namespace.h
+ */
+extern int is_global_init(struct task_struct *tsk);
+
+/*
+ * is_container_init:
+ * check whether in the task is init in its own pid namespace.
  */
-static inline int is_init(struct task_struct *tsk)
+static inline int is_container_init(struct task_struct *tsk)
 {
         return tsk->pid == 1;
 }
diff --git a/kernel/capability.c b/kernel/capability.c
index cbc5fd60c0f3..f02ad47320b9 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -12,6 +12,7 @@
 #include <linux/module.h>
 #include <linux/security.h>
 #include <linux/syscalls.h>
+#include <linux/pid_namespace.h>
 #include <asm/uaccess.h>
 
 /*
@@ -129,7 +130,7 @@ static inline int cap_set_all(kernel_cap_t *effective,
      int found = 0;
 
      do_each_thread(g, target) {
-             if (target == current || is_init(target))
+             if (target == current || is_container_init(target->group_leader))
                      continue;
              found = 1;
              if (security_capset_check(target, effective, inheritable,
diff --git a/kernel/exit.c b/kernel/exit.c
index d1eddc753fe3..d22aefabb129 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -221,7 +221,7 @@ static int will_become_orphaned_pgrp(struct pid *pgrp, struct task_struct *ignor
         do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
                 if (p == ignored_task
                                 || p->exit_state
-                                || is_init(p->real_parent))
+                                || is_global_init(p->real_parent))
                         continue;
                 if (task_pgrp(p->real_parent) != pgrp &&
                     task_session(p->real_parent) == task_session(p)) {
diff --git a/kernel/kexec.c b/kernel/kexec.c
index e9f1b4ea504d..fbffdb457cce 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -51,7 +51,7 @@ struct resource crashk_res = {
 
 int kexec_should_crash(struct task_struct *p)
 {
-        if (in_interrupt() || !p->pid || is_init(p) || panic_on_oops)
+        if (in_interrupt() || !p->pid || is_global_init(p) || panic_on_oops)
                 return 1;
         return 0;
 }
diff --git a/kernel/pid.c b/kernel/pid.c
index 78c0dbffde65..bb0785109d39 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -70,6 +70,11 @@ struct pid_namespace init_pid_ns = {
         .child_reaper = &init_task
 };
 
+int is_global_init(struct task_struct *tsk)
+{
+        return tsk == init_pid_ns.child_reaper;
+}
+
 /*
  * Note: disable interrupts while the pidmap_lock is held as an
  * interrupt might come in and do read_lock(&tasklist_lock).
diff --git a/kernel/signal.c b/kernel/signal.c
index 0a6d3726cb80..8214ffad54bc 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -256,7 +256,7 @@ flush_signal_handlers(struct task_struct *t, int force_default)
 
 int unhandled_signal(struct task_struct *tsk, int sig)
 {
-        if (is_init(tsk))
+        if (is_global_init(tsk))
                 return 1;
         if (tsk->ptrace & PT_PTRACED)
                 return 0;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 067554bda8b7..44868e4df1d3 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1888,7 +1888,7 @@ int proc_dointvec_bset(struct ctl_table *table, int write, struct file *filp,
                 return -EPERM;
         }
 
-        op = is_init(current) ? OP_SET : OP_AND;
+        op = is_global_init(current) ? OP_SET : OP_AND;
         return do_proc_dointvec(table,write,filp,buffer,lenp,ppos,
                                 do_proc_dointvec_bset_conv,&op);
 }
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index a64decb5b13f..b1c2d0f86222 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -212,7 +212,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
                 if (!p->mm)
                         continue;
                 /* skip the init task */
-                if (is_init(p))
+                if (is_global_init(p))
                         continue;
 
                 /*
@@ -265,7 +265,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
  */
 static void __oom_kill_task(struct task_struct *p, int verbose)
 {
-        if (is_init(p)) {
+        if (is_global_init(p)) {
                 WARN_ON(1);
                 printk(KERN_WARNING "tried to kill init!\n");
                 return;
diff --git a/security/commoncap.c b/security/commoncap.c
index 48ca5b092768..43f902750a1b 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -23,6 +23,7 @@
 #include <linux/xattr.h>
 #include <linux/hugetlb.h>
 #include <linux/mount.h>
+#include <linux/sched.h>
 
 #ifdef CONFIG_SECURITY_FILE_CAPABILITIES
 /*
@@ -334,7 +335,7 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
         /* For init, we want to retain the capabilities set
          * in the init_task struct. Thus we skip the usual
          * capability rules */
-        if (!is_init(current)) {
+        if (!is_global_init(current)) {
                 current->cap_permitted = new_permitted;
                 current->cap_effective = bprm->cap_effective ?
                                 new_permitted : 0;