From: sukadev@linux.vnet.ibm.com
Subject: [PATCH 17/17] Pid-NS(V3) Introduce proc_mnt for pid_ns

The following patch completes the removal of the global proc_mnt.
It fetches the mnt on which to do dentry invalidations from the
pid_namespace in which the task appears.

For now, there is only one pid namespace in mainline so this is
straightforward.  In the -lxc tree we'll have to do something
more complex.  The proc_flush_task() code takes a task, and
needs to be able to find the corresponding proc superblocks on
which that tasks's /proc/<pid> directories could appear.  We
can tell in which pid namespaces a task appears, so I put a
pointer from the pid namespace to the corresponding proc_mnt.

/proc currently has some special code to make sure that the root
directory gets set up correctly.  It proc_mnt variable in order
to find its way to the root inode.

Changelog:

	2.6.22-rc4-mm2-pidns1:

	- Call proc_fill_super once per pid namespace
	- Call proc_flush_task() before detaching a task's 'struct pid'.
	- Get a reference to pid namespace when mounting/remounting /proc.
	  Put this reference when unmounting.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
---

 fs/proc/base.c                |   30 ++++++++++++++++++-
 fs/proc/inode.c               |   12 ++++++-
 fs/proc/root.c                |   65 ++++++++++++++++++++++++++++++------------
 include/linux/pid_namespace.h |    1 
 include/linux/proc_fs.h       |    1 
 kernel/exit.c                 |    2 -
 6 files changed, 87 insertions(+), 24 deletions(-)

Index: lx26-22-rc4-mm2/fs/proc/base.c
===================================================================
--- lx26-22-rc4-mm2.orig/fs/proc/base.c	2007-06-16 02:37:15.000000000 -0700
+++ lx26-22-rc4-mm2/fs/proc/base.c	2007-06-16 04:15:23.000000000 -0700
@@ -2164,9 +2164,11 @@ static const struct inode_operations pro
 };
 
 /**
- * proc_flush_task -  Remove dcache entries for @task from the /proc dcache.
+ * proc_flush_task_from_pid_ns -  Remove dcache entries for @task
+ * 				  from the /proc dcache.
  *
  * @task: task that should be flushed.
+ * @pid_ns: pid_namespace in which that task appears
  *
  * Looks in the dcache for
  * /proc/@pid
@@ -2184,11 +2186,20 @@ static const struct inode_operations pro
  *       that no dcache entries will exist at process exit time it
  *       just makes it very unlikely that any will persist.
  */
-void proc_flush_task(struct task_struct *task)
+static void proc_flush_task_from_pid_ns(struct task_struct *task,
+				struct pid_namespace* pid_ns)
 {
 	struct dentry *dentry, *leader, *dir;
 	char buf[PROC_NUMBUF];
 	struct qstr name;
+	struct vfsmount *proc_mnt = pid_ns->proc_mnt;
+
+	/*
+	 * It is possible that no /procs have been instantiated
+	 * for this particular pid namespace.
+	 */
+	if (!proc_mnt || !proc_mnt->mnt_root)
+		return;
 
 	name.name = buf;
 	name.len = snprintf(buf, sizeof(buf), "%d", task->pid);
@@ -2230,6 +2241,21 @@ out:
 	return;
 }
 
+void proc_flush_task(struct task_struct *task)
+{
+	int i;
+	struct pid *pid;
+	struct upid* upid;
+
+	pid = task_pid(task);
+	if (!pid)
+		return;
+
+	upid = &pid->upid_list[0];
+	for (i = 0; i < pid->num_upids; i++, upid++)
+		proc_flush_task_from_pid_ns(task, upid->pid_ns);
+}
+
 static struct dentry *proc_pid_instantiate(struct inode *dir,
 					   struct dentry * dentry,
 					   struct task_struct *task, const void *ptr)
Index: lx26-22-rc4-mm2/fs/proc/inode.c
===================================================================
--- lx26-22-rc4-mm2.orig/fs/proc/inode.c	2007-06-16 02:37:15.000000000 -0700
+++ lx26-22-rc4-mm2/fs/proc/inode.c	2007-06-16 04:15:23.000000000 -0700
@@ -6,6 +6,7 @@
 
 #include <linux/time.h>
 #include <linux/proc_fs.h>
+#include <linux/hardirq.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
 #include <linux/string.h>
@@ -19,6 +20,7 @@
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
+#include <linux/pid_namespace.h>
 
 #include "internal.h"
 
@@ -75,8 +77,6 @@ static void proc_delete_inode(struct ino
 	clear_inode(inode);
 }
 
-struct vfsmount *proc_mnt;
-
 static void proc_read_inode(struct inode * inode)
 {
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
@@ -431,6 +431,8 @@ out_mod:
 
 int proc_fill_super(struct super_block *s, void *data, int silent)
 {
+	struct pid_namespace *pid_ns = data;
+	struct proc_inode *ei;
 	struct inode * root_inode;
 
 	s->s_flags |= MS_NODIRATIME | MS_NOSUID | MS_NOEXEC;
@@ -439,6 +441,7 @@ int proc_fill_super(struct super_block *
 	s->s_magic = PROC_SUPER_MAGIC;
 	s->s_op = &proc_sops;
 	s->s_time_gran = 1;
+	s->s_fs_info = get_pid_ns(pid_ns);
 	
 	de_get(&proc_root);
 	root_inode = proc_get_inode(s, PROC_ROOT_INO, &proc_root);
@@ -449,6 +452,11 @@ int proc_fill_super(struct super_block *
 	s->s_root = d_alloc_root(root_inode);
 	if (!s->s_root)
 		goto out_no_root;
+	/* Seed the root directory with a pid so it doesn't need
+	 * to be special in base.c.
+	 */
+	ei = PROC_I(root_inode);
+	ei->pid = find_get_pid(1);
 	return 0;
 
 out_no_root:
Index: lx26-22-rc4-mm2/fs/proc/root.c
===================================================================
--- lx26-22-rc4-mm2.orig/fs/proc/root.c	2007-06-16 04:15:23.000000000 -0700
+++ lx26-22-rc4-mm2/fs/proc/root.c	2007-06-16 04:15:23.000000000 -0700
@@ -12,38 +12,73 @@
 #include <linux/time.h>
 #include <linux/proc_fs.h>
 #include <linux/stat.h>
+#include <linux/hardirq.h>
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/module.h>
 #include <linux/bitops.h>
 #include <linux/smp_lock.h>
 #include <linux/mount.h>
+#include <linux/pid_namespace.h>
 
 #include "internal.h"
 
 struct proc_dir_entry *proc_net, *proc_net_stat, *proc_bus, *proc_root_fs, *proc_root_driver;
 
+static int proc_test_sb(struct super_block *s, void *data)
+{
+	struct pid_namespace *pid_ns = data;
+	if (s->s_fs_info == pid_ns)
+		return 1;
+	return 0;
+}
+
 static int proc_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *data, struct vfsmount *mnt)
 {
-	if (proc_mnt) {
-		/* Seed the root directory with a pid so it doesn't need
-		 * to be special in base.c.  I would do this earlier but
-		 * the only task alive when /proc is mounted the first time
-		 * is the init_task and it doesn't have any pids.
-		 */
-		struct proc_inode *ei;
-		ei = PROC_I(proc_mnt->mnt_sb->s_root->d_inode);
-		if (!ei->pid)
-			ei->pid = find_get_pid(1);
+	int error;
+	struct super_block *s;
+	struct pid_namespace *pid_ns;
+
+	/*
+	 * We can eventually derive this out of whatever mount
+	 * arguments the user supplies, but just take it from
+	 * current for now.
+	 */
+	pid_ns = task_active_pid_ns(current);
+
+	s = sget(fs_type, proc_test_sb, set_anon_super, pid_ns);
+	if (IS_ERR(s))
+		return PTR_ERR(s);
+
+	if (!s->s_root) {
+		error = proc_fill_super(s, pid_ns, 0);
+		if (error) {
+			deactivate_super(s);
+			return error;
+		}
 	}
-	return get_sb_single(fs_type, flags, data, proc_fill_super, mnt);
+
+	if (!pid_ns->proc_mnt)
+		pid_ns->proc_mnt = mnt;
+
+	do_remount_sb(s, flags, data, 0);
+	return simple_set_mnt(mnt, s);
+}
+
+static void proc_kill_sb(struct super_block *s)
+{
+	struct pid_namespace *pid_ns = s->s_fs_info;
+
+	pid_ns->proc_mnt = NULL;
+	put_pid_ns(pid_ns);
+	kill_anon_super(s);
 }
 
 static struct file_system_type proc_fs_type = {
 	.name		= "proc",
 	.get_sb		= proc_get_sb,
-	.kill_sb	= kill_anon_super,
+	.kill_sb	= proc_kill_sb,
 	.fs_supers      = LIST_HEAD_INIT(proc_fs_type.fs_supers),
 };
 
@@ -55,12 +90,6 @@ void __init proc_root_init(void)
 	err = register_filesystem(&proc_fs_type);
 	if (err)
 		return;
-	proc_mnt = kern_mount(&proc_fs_type);
-	err = PTR_ERR(proc_mnt);
-	if (IS_ERR(proc_mnt)) {
-		unregister_filesystem(&proc_fs_type);
-		return;
-	}
 	proc_misc_init();
 	proc_net = proc_mkdir("net", NULL);
 	proc_net_stat = proc_mkdir("net/stat", NULL);
Index: lx26-22-rc4-mm2/include/linux/pid_namespace.h
===================================================================
--- lx26-22-rc4-mm2.orig/include/linux/pid_namespace.h	2007-06-16 04:15:23.000000000 -0700
+++ lx26-22-rc4-mm2/include/linux/pid_namespace.h	2007-06-16 04:15:23.000000000 -0700
@@ -28,6 +28,7 @@ struct pid_namespace {
 	int last_pid;
 	struct task_struct *child_reaper;
 	atomic_t terminating;
+	struct vfsmount *proc_mnt;
 };
 
 extern struct pid_namespace init_pid_ns;
Index: lx26-22-rc4-mm2/include/linux/proc_fs.h
===================================================================
--- lx26-22-rc4-mm2.orig/include/linux/proc_fs.h	2007-06-16 02:37:15.000000000 -0700
+++ lx26-22-rc4-mm2/include/linux/proc_fs.h	2007-06-16 04:15:23.000000000 -0700
@@ -125,7 +125,6 @@ extern struct proc_dir_entry *create_pro
 						struct proc_dir_entry *parent);
 extern void remove_proc_entry(const char *name, struct proc_dir_entry *parent);
 
-extern struct vfsmount *proc_mnt;
 extern int proc_fill_super(struct super_block *,void *,int);
 extern struct inode *proc_get_inode(struct super_block *, unsigned int, struct proc_dir_entry *);
 
Index: lx26-22-rc4-mm2/kernel/exit.c
===================================================================
--- lx26-22-rc4-mm2.orig/kernel/exit.c	2007-06-16 04:15:23.000000000 -0700
+++ lx26-22-rc4-mm2/kernel/exit.c	2007-06-16 04:15:23.000000000 -0700
@@ -157,6 +157,7 @@ void release_task(struct task_struct * p
 	struct task_struct *leader;
 	int zap_leader;
 repeat:
+	proc_flush_task(p);
 	atomic_dec(&p->user->processes);
 	write_lock_irq(&tasklist_lock);
 	ptrace_unlink(p);
@@ -185,7 +186,6 @@ repeat:
 	}
 
 	write_unlock_irq(&tasklist_lock);
-	proc_flush_task(p);
 	release_thread(p);
 	call_rcu(&am

sukadev@us.ibm.com wrote:
> From: sukadev@linux.vnet.ibm.com
> Subject: [PATCH 17/17] Pid-NS(V3) Introduce proc_mnt for pid_ns
> 
> The following patch completes the removal of the global proc_mnt.
> It fetches the mnt on which to do dentry invalidations from the
> pid_namespace in which the task appears.
> 
> For now, there is only one pid namespace in mainline so this is
> straightforward.  In the -lxc tree we'll have to do something
> more complex.  The proc_flush_task() code takes a task, and
> needs to be able to find the corresponding proc superblocks on
> which that tasks's /proc/<pid> directories could appear.  We
> can tell in which pid namespaces a task appears, so I put a
> pointer from the pid namespace to the corresponding proc_mnt.
> 
> /proc currently has some special code to make sure that the root
> directory gets set up correctly.  It proc_mnt variable in order
> to find its way to the root inode.
> 
> Changelog:
> 
> 	2.6.22-rc4-mm2-pidns1:
> 
> 	- Call proc_fill_super once per pid namespace
> 	- Call proc_flush_task() before detaching a task's 'struct pid'.
> 	- Get a reference to pid namespace when mounting/remounting /proc.
> 	  Put this reference when unmounting.
> 
> Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
> Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
> ---
> 
>  fs/proc/base.c                |   30 ++++++++++++++++++-
>  fs/proc/inode.c               |   12 ++++++-
>  fs/proc/root.c                |   65 ++++++++++++++++++++++++++++++------------
>  include/linux/pid_namespace.h |    1 
>  include/linux/proc_fs.h       |    1 
>  kernel/exit.c                 |    2 -
>  6 files changed, 87 insertions(+), 24 deletions(-)
> 
[snip]

> @@ -55,12 +90,6 @@ void __init proc_root_init(void)
>  	err = register_filesystem(&proc_fs_type);
>  	if (err)
>  		return;
> -	proc_mnt = kern_mount(&proc_fs_type);
> -	err = PTR_ERR(proc_mnt);
> -	if (IS_ERR(proc_mnt)) {
> -		unregister_filesystem(&proc_fs_type);
> -		return;
> -	}

Wow! Is this safe? Everyone expexts proc_mnt to be not NULL.

>  	proc_misc_init();
>  	proc_net = proc_mkdir("net", NULL);
>  	proc_net_stat = proc_mkdir("net/stat", NULL);

[snip]

> Index: lx26-22-rc4-mm2/kernel/exit.c
> ===================================================================
> --- lx26-22-rc4-mm2.orig/kernel/exit.c	2007-06-16 04:15:23.000000000 -0700
> +++ lx26-22-rc4-mm2/kernel/exit.c	2007-06-16 04:15:23.000000000 -0700
> @@ -157,6 +157,7 @@ void release_task(struct task_struct * p
>  	struct task_struct *leader;
>  	int zap_leader;
>  repeat:
> +	proc_flush_task(p);

Such a flush won't actually remove *any* dentries from the tree and thus
may be omitted.

>  	atomic_dec(&p->user->processes);
>  	write_lock_irq(&tasklist_lock);
>  	ptrace_unlink(p);
> @@ -185,7 +186,6 @@ repeat:
>  	}
>  
>  	write_unlock_irq(&tasklist_lock);
> -	proc_flush_task(p);
>  	release_thread(p);
>  	call_rcu(&p->rcu, delayed_put_task_struct);
>  
> 

_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers

Pavel Emelianov [xemul@openvz.org] wrote:
| sukadev@us.ibm.com wrote:
| > From: sukadev@linux.vnet.ibm.com
| > Subject: [PATCH 17/17] Pid-NS(V3) Introduce proc_mnt for pid_ns
| > 
| > The following patch completes the removal of the global proc_mnt.
| > It fetches the mnt on which to do dentry invalidations from the
| > pid_namespace in which the task appears.
| > 
| > For now, there is only one pid namespace in mainline so this is
| > straightforward.  In the -lxc tree we'll have to do something
| > more complex.  The proc_flush_task() code takes a task, and
| > needs to be able to find the corresponding proc superblocks on
| > which that tasks's /proc/<pid> directories could appear.  We
| > can tell in which pid namespaces a task appears, so I put a
| > pointer from the pid namespace to the corresponding proc_mnt.
| > 
| > /proc currently has some special code to make sure that the root
| > directory gets set up correctly.  It proc_mnt variable in order
| > to find its way to the root inode.
| > 
| > Changelog:
| > 
| > 	2.6.22-rc4-mm2-pidns1:
| > 
| > 	- Call proc_fill_super once per pid namespace
| > 	- Call proc_flush_task() before detaching a task's 'struct pid'.
| > 	- Get a reference to pid namespace when mounting/remounting /proc.
| > 	  Put this reference when unmounting.
| > 
| > Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
| > Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
| > ---
| > 
| >  fs/proc/base.c                |   30 ++++++++++++++++++-
| >  fs/proc/inode.c               |   12 ++++++-
| >  fs/proc/root.c                |   65 ++++++++++++++++++++++++++++++------------
| >  include/linux/pid_namespace.h |    1 
| >  include/linux/proc_fs.h       |    1 
| >  kernel/exit.c                 |    2 -
| >  6 files changed, 87 insertions(+), 24 deletions(-)
| > 
| [snip]
| 
| > @@ -55,12 +90,6 @@ void __init proc_root_init(void)
| >  	err = register_filesystem(&proc_fs_type);
| >  	if (err)
| >  		return;
| > -	proc_mnt = kern_mount(&proc_fs_type);
| > -	err = PTR_ERR(proc_mnt);
| > -	if (IS_ERR(proc_mnt)) {
| > -		unregister_filesystem(&proc_fs_type);
| > -		return;
| > -	}
| 
| Wow! Is this safe? Everyone expexts proc_mnt to be not NULL.

Can you elaborate ? Between the previous patch (i.e 16/17) and this
we made sure no one is using the global proc_mnt. Everyone should be
using the pid_ns->proc_mnt instead.

| 
| >  	proc_misc_init();
| >  	proc_net = proc_mkdir("net", NULL);
| >  	proc_net_stat = proc_mkdir("net/stat", NULL);
| 
| [snip]
| 
| > Index: lx26-22-rc4-mm2/kernel/exit.c
| > ===================================================================
| > --- lx26-22-rc4-mm2.orig/kernel/exit.c	2007-06-16 04:15:23.000000000 -0700
| > +++ lx26-22-rc4-mm2/kernel/exit.c	2007-06-16 04:15:23.000000000 -0700
| > @@ -157,6 +157,7 @@ void release_task(struct task_struct * p
| >  	struct task_struct *leader;
| >  	int zap_leader;
| >  repeat:
| > +	proc_flush_task(p);
| 
| Such a flush won't actually remove *any* dentries from the tree and thus
| may be omitted.

Again, can you elaborate ?

The previous version had proc_flush_task() much later in release_task() and
that did not free dentries bc task_pid() was NULL. Moving proc_flush_task()
up, should fix that and it did fix the leak we were seeing before.

| 
| >  	atomic_dec(&p->user->processes);
| >  	write_lock_irq(&tasklist_lock);
| >  	ptrace_unlink(p);
| > @@ -185,7 +186,6 @@ repeat:
| >  	}
| >  
| >  	write_unlock_irq(&tasklist_lock);
| > -	proc_flush_task(p);
| >  	release_thread(p);
| >  	call_rcu(&p->rcu, delayed_put_task_struct);
| >  
| > 
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers