From 49f4d8b93ccf9454284b6f524b96c66d8d7fbccc Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Thu, 2 Aug 2012 04:25:10 -0700 Subject: pidns: Capture the user namespace and filter ns_last_pid - Capture the the user namespace that creates the pid namespace - Use that user namespace to test if it is ok to write to /proc/sys/kernel/ns_last_pid. Zhao Hongjiang noticed I was missing a put_user_ns in when destroying a pid_ns. I have foloded his patch into this one so that bisects will work properly. Acked-by: Serge Hallyn Signed-off-by: "Eric W. Biederman" --- include/linux/pid_namespace.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'include/linux/pid_namespace.h') diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index 65e3e87eacc..c89c9cfcd24 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -31,6 +31,7 @@ struct pid_namespace { #ifdef CONFIG_BSD_PROCESS_ACCT struct bsd_acct_struct *bacct; #endif + struct user_namespace *user_ns; kgid_t pid_gid; int hide_pid; int reboot; /* group exit code if this pidns was rebooted */ @@ -46,7 +47,8 @@ static inline struct pid_namespace *get_pid_ns(struct pid_namespace *ns) return ns; } -extern struct pid_namespace *copy_pid_ns(unsigned long flags, struct pid_namespace *ns); +extern struct pid_namespace *copy_pid_ns(unsigned long flags, + struct user_namespace *user_ns, struct pid_namespace *ns); extern void zap_pid_ns_processes(struct pid_namespace *pid_ns); extern int reboot_pid_ns(struct pid_namespace *pid_ns, int cmd); extern void put_pid_ns(struct pid_namespace *ns); @@ -59,8 +61,8 @@ static inline struct pid_namespace *get_pid_ns(struct pid_namespace *ns) return ns; } -static inline struct pid_namespace * -copy_pid_ns(unsigned long flags, struct pid_namespace *ns) +static inline struct pid_namespace *copy_pid_ns(unsigned long flags, + struct user_namespace *user_ns, struct pid_namespace *ns) { if (flags & CLONE_NEWPID) ns = ERR_PTR(-EINVAL); -- cgit v1.2.3-70-g09d2 From 0a01f2cc390e10633a54f72c608cc3fe19a50c3d Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 1 Aug 2012 10:33:47 -0700 Subject: pidns: Make the pidns proc mount/umount logic obvious. Track the number of pids in the proc hash table. When the number of pids goes to 0 schedule work to unmount the kernel mount of proc. Move the mount of proc into alloc_pid when we allocate the pid for init. Remove the surprising calls of pid_ns_release proc in fork and proc_flush_task. Those code paths really shouldn't know about proc namespace implementation details and people have demonstrated several times that finding and understanding those code paths is difficult and non-obvious. Because of the call path detach pid is alwasy called with the rtnl_lock held free_pid is not allowed to sleep, so the work to unmounting proc is moved to a work queue. This has the side benefit of not blocking the entire world waiting for the unnecessary rcu_barrier in deactivate_locked_super. In the process of making the code clear and obvious this fixes a bug reported by Gao feng where we would leak a mount of proc during clone(CLONE_NEWPID|CLONE_NEWNET) if copy_pid_ns succeeded and copy_net_ns failed. Acked-by: "Serge E. Hallyn" Signed-off-by: "Eric W. Biederman" --- fs/proc/base.c | 4 ---- fs/proc/root.c | 5 ----- include/linux/pid_namespace.h | 2 ++ kernel/fork.c | 2 -- kernel/pid.c | 21 +++++++++++++++++---- kernel/pid_namespace.c | 14 +++++++------- 6 files changed, 26 insertions(+), 22 deletions(-) (limited to 'include/linux/pid_namespace.h') diff --git a/fs/proc/base.c b/fs/proc/base.c index 6177fc238fd..7621dc51cff 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2590,10 +2590,6 @@ void proc_flush_task(struct task_struct *task) proc_flush_task_mnt(upid->ns->proc_mnt, upid->nr, tgid->numbers[i].nr); } - - upid = &pid->numbers[pid->level]; - if (upid->nr == 1) - pid_ns_release_proc(upid->ns); } static struct dentry *proc_pid_instantiate(struct inode *dir, diff --git a/fs/proc/root.c b/fs/proc/root.c index fc1609321a7..f2f251158d3 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -155,11 +155,6 @@ void __init proc_root_init(void) err = register_filesystem(&proc_fs_type); if (err) return; - err = pid_ns_prepare_proc(&init_pid_ns); - if (err) { - unregister_filesystem(&proc_fs_type); - return; - } proc_self_init(); proc_symlink("mounts", NULL, "self/mounts"); diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index c89c9cfcd24..4c96acdb248 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -21,6 +21,7 @@ struct pid_namespace { struct kref kref; struct pidmap pidmap[PIDMAP_ENTRIES]; int last_pid; + int nr_hashed; struct task_struct *child_reaper; struct kmem_cache *pid_cachep; unsigned int level; @@ -32,6 +33,7 @@ struct pid_namespace { struct bsd_acct_struct *bacct; #endif struct user_namespace *user_ns; + struct work_struct proc_work; kgid_t pid_gid; int hide_pid; int reboot; /* group exit code if this pidns was rebooted */ diff --git a/kernel/fork.c b/kernel/fork.c index 7798c247f4b..666dc8b0660 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1476,8 +1476,6 @@ bad_fork_cleanup_io: if (p->io_context) exit_io_context(p); bad_fork_cleanup_namespaces: - if (unlikely(clone_flags & CLONE_NEWPID)) - pid_ns_release_proc(p->nsproxy->pid_ns); exit_task_namespaces(p); bad_fork_cleanup_mm: if (p->mm) diff --git a/kernel/pid.c b/kernel/pid.c index 3a5f238c1ca..e957f8b0913 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -36,6 +36,7 @@ #include #include #include +#include #define pid_hashfn(nr, ns) \ hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift) @@ -270,8 +271,12 @@ void free_pid(struct pid *pid) unsigned long flags; spin_lock_irqsave(&pidmap_lock, flags); - for (i = 0; i <= pid->level; i++) - hlist_del_rcu(&pid->numbers[i].pid_chain); + for (i = 0; i <= pid->level; i++) { + struct upid *upid = pid->numbers + i; + hlist_del_rcu(&upid->pid_chain); + if (--upid->ns->nr_hashed == 0) + schedule_work(&upid->ns->proc_work); + } spin_unlock_irqrestore(&pidmap_lock, flags); for (i = 0; i <= pid->level; i++) @@ -293,6 +298,7 @@ struct pid *alloc_pid(struct pid_namespace *ns) goto out; tmp = ns; + pid->level = ns->level; for (i = ns->level; i >= 0; i--) { nr = alloc_pidmap(tmp); if (nr < 0) @@ -303,17 +309,23 @@ struct pid *alloc_pid(struct pid_namespace *ns) tmp = tmp->parent; } + if (unlikely(is_child_reaper(pid))) { + if (pid_ns_prepare_proc(ns)) + goto out_free; + } + get_pid_ns(ns); - pid->level = ns->level; atomic_set(&pid->count, 1); for (type = 0; type < PIDTYPE_MAX; ++type) INIT_HLIST_HEAD(&pid->tasks[type]); upid = pid->numbers + ns->level; spin_lock_irq(&pidmap_lock); - for ( ; upid >= pid->numbers; --upid) + for ( ; upid >= pid->numbers; --upid) { hlist_add_head_rcu(&upid->pid_chain, &pid_hash[pid_hashfn(upid->nr, upid->ns)]); + upid->ns->nr_hashed++; + } spin_unlock_irq(&pidmap_lock); out: @@ -570,6 +582,7 @@ void __init pidmap_init(void) /* Reserve PID 0. We never call free_pidmap(0) */ set_bit(0, init_pid_ns.pidmap[0].page); atomic_dec(&init_pid_ns.pidmap[0].nr_free); + init_pid_ns.nr_hashed = 1; init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC); diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index b2604950aa5..84591cfeefc 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -72,6 +72,12 @@ err_alloc: return NULL; } +static void proc_cleanup_work(struct work_struct *work) +{ + struct pid_namespace *ns = container_of(work, struct pid_namespace, proc_work); + pid_ns_release_proc(ns); +} + /* MAX_PID_NS_LEVEL is needed for limiting size of 'struct pid' */ #define MAX_PID_NS_LEVEL 32 @@ -105,6 +111,7 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns ns->level = level; ns->parent = get_pid_ns(parent_pid_ns); ns->user_ns = get_user_ns(user_ns); + INIT_WORK(&ns->proc_work, proc_cleanup_work); set_bit(0, ns->pidmap[0].page); atomic_set(&ns->pidmap[0].nr_free, BITS_PER_PAGE - 1); @@ -112,15 +119,8 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns for (i = 1; i < PIDMAP_ENTRIES; i++) atomic_set(&ns->pidmap[i].nr_free, BITS_PER_PAGE); - err = pid_ns_prepare_proc(ns); - if (err) - goto out_put_parent_pid_ns; - return ns; -out_put_parent_pid_ns: - put_pid_ns(parent_pid_ns); - put_user_ns(user_ns); out_free_map: kfree(ns->pidmap[0].page); out_free: -- cgit v1.2.3-70-g09d2 From 98f842e675f96ffac96e6c50315790912b2812be Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 15 Jun 2011 10:21:48 -0700 Subject: proc: Usable inode numbers for the namespace file descriptors. Assign a unique proc inode to each namespace, and use that inode number to ensure we only allocate at most one proc inode for every namespace in proc. A single proc inode per namespace allows userspace to test to see if two processes are in the same namespace. This has been a long requested feature and only blocked because a naive implementation would put the id in a global space and would ultimately require having a namespace for the names of namespaces, making migration and certain virtualization tricks impossible. We still don't have per superblock inode numbers for proc, which appears necessary for application unaware checkpoint/restart and migrations (if the application is using namespace file descriptors) but that is now allowd by the design if it becomes important. I have preallocated the ipc and uts initial proc inode numbers so their structures can be statically initialized. Signed-off-by: Eric W. Biederman --- fs/mount.h | 1 + fs/namespace.c | 14 ++++++++++++++ fs/proc/namespaces.c | 24 ++++++++++++++---------- include/linux/ipc_namespace.h | 2 ++ include/linux/pid_namespace.h | 1 + include/linux/proc_fs.h | 7 ++++++- include/linux/user_namespace.h | 1 + include/linux/utsname.h | 1 + include/net/net_namespace.h | 2 ++ init/version.c | 2 ++ ipc/msgutil.c | 2 ++ ipc/namespace.c | 16 ++++++++++++++++ kernel/pid.c | 1 + kernel/pid_namespace.c | 12 ++++++++++++ kernel/user.c | 2 ++ kernel/user_namespace.c | 15 +++++++++++++++ kernel/utsname.c | 17 ++++++++++++++++- net/core/net_namespace.c | 24 ++++++++++++++++++++++++ 18 files changed, 132 insertions(+), 12 deletions(-) (limited to 'include/linux/pid_namespace.h') diff --git a/fs/mount.h b/fs/mount.h index 630fafc616b..cd500798040 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -4,6 +4,7 @@ struct mnt_namespace { atomic_t count; + unsigned int proc_inum; struct mount * root; struct list_head list; struct user_namespace *user_ns; diff --git a/fs/namespace.c b/fs/namespace.c index cab78a74aca..c1bbe86f492 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2301,6 +2301,7 @@ dput_out: static void free_mnt_ns(struct mnt_namespace *ns) { + proc_free_inum(ns->proc_inum); put_user_ns(ns->user_ns); kfree(ns); } @@ -2317,10 +2318,16 @@ static atomic64_t mnt_ns_seq = ATOMIC64_INIT(1); static struct mnt_namespace *alloc_mnt_ns(struct user_namespace *user_ns) { struct mnt_namespace *new_ns; + int ret; new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL); if (!new_ns) return ERR_PTR(-ENOMEM); + ret = proc_alloc_inum(&new_ns->proc_inum); + if (ret) { + kfree(new_ns); + return ERR_PTR(ret); + } new_ns->seq = atomic64_add_return(1, &mnt_ns_seq); atomic_set(&new_ns->count, 1); new_ns->root = NULL; @@ -2799,10 +2806,17 @@ static int mntns_install(struct nsproxy *nsproxy, void *ns) return 0; } +static unsigned int mntns_inum(void *ns) +{ + struct mnt_namespace *mnt_ns = ns; + return mnt_ns->proc_inum; +} + const struct proc_ns_operations mntns_operations = { .name = "mnt", .type = CLONE_NEWNS, .get = mntns_get, .put = mntns_put, .install = mntns_install, + .inum = mntns_inum, }; diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index 7a6d8d69cdb..b7a47196c8c 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -82,7 +82,7 @@ static struct dentry *proc_ns_get_dentry(struct super_block *sb, return ERR_PTR(-ENOMEM); } - inode = new_inode(sb); + inode = iget_locked(sb, ns_ops->inum(ns)); if (!inode) { dput(dentry); ns_ops->put(ns); @@ -90,13 +90,17 @@ static struct dentry *proc_ns_get_dentry(struct super_block *sb, } ei = PROC_I(inode); - inode->i_ino = get_next_ino(); - inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_op = &ns_inode_operations; - inode->i_mode = S_IFREG | S_IRUGO; - inode->i_fop = &ns_file_operations; - ei->ns_ops = ns_ops; - ei->ns = ns; + if (inode->i_state & I_NEW) { + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; + inode->i_op = &ns_inode_operations; + inode->i_mode = S_IFREG | S_IRUGO; + inode->i_fop = &ns_file_operations; + ei->ns_ops = ns_ops; + ei->ns = ns; + unlock_new_inode(inode); + } else { + ns_ops->put(ns); + } d_set_d_op(dentry, &ns_dentry_operations); result = d_instantiate_unique(dentry, inode); @@ -162,12 +166,12 @@ static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int bufl if (!ns) goto out_put_task; - snprintf(name, sizeof(name), "%s", ns_ops->name); + snprintf(name, sizeof(name), "%s:[%u]", ns_ops->name, ns_ops->inum(ns)); len = strlen(name); if (len > buflen) len = buflen; - if (copy_to_user(buffer, ns_ops->name, len)) + if (copy_to_user(buffer, name, len)) len = -EFAULT; ns_ops->put(ns); diff --git a/include/linux/ipc_namespace.h b/include/linux/ipc_namespace.h index f03af702a39..fe771978e87 100644 --- a/include/linux/ipc_namespace.h +++ b/include/linux/ipc_namespace.h @@ -67,6 +67,8 @@ struct ipc_namespace { /* user_ns which owns the ipc ns */ struct user_namespace *user_ns; + + unsigned int proc_inum; }; extern struct ipc_namespace init_ipc_ns; diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index 4c96acdb248..bf285999273 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -37,6 +37,7 @@ struct pid_namespace { kgid_t pid_gid; int hide_pid; int reboot; /* group exit code if this pidns was rebooted */ + unsigned int proc_inum; }; extern struct pid_namespace init_pid_ns; diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index bf1d000fbba..2e24018b7ce 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -28,7 +28,11 @@ struct mm_struct; */ enum { - PROC_ROOT_INO = 1, + PROC_ROOT_INO = 1, + PROC_IPC_INIT_INO = 0xEFFFFFFFU, + PROC_UTS_INIT_INO = 0xEFFFFFFEU, + PROC_USER_INIT_INO = 0xEFFFFFFDU, + PROC_PID_INIT_INO = 0xEFFFFFFCU, }; /* @@ -263,6 +267,7 @@ struct proc_ns_operations { void *(*get)(struct task_struct *task); void (*put)(void *ns); int (*install)(struct nsproxy *nsproxy, void *ns); + unsigned int (*inum)(void *ns); }; extern const struct proc_ns_operations netns_operations; extern const struct proc_ns_operations utsns_operations; diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h index 17651f08d67..b9bd2e6c73c 100644 --- a/include/linux/user_namespace.h +++ b/include/linux/user_namespace.h @@ -25,6 +25,7 @@ struct user_namespace { struct user_namespace *parent; kuid_t owner; kgid_t group; + unsigned int proc_inum; }; extern struct user_namespace init_user_ns; diff --git a/include/linux/utsname.h b/include/linux/utsname.h index 221f4a0a750..239e27733d6 100644 --- a/include/linux/utsname.h +++ b/include/linux/utsname.h @@ -23,6 +23,7 @@ struct uts_namespace { struct kref kref; struct new_utsname name; struct user_namespace *user_ns; + unsigned int proc_inum; }; extern struct uts_namespace init_uts_ns; diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index c5a43f56b79..de644bcd861 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -56,6 +56,8 @@ struct net { struct user_namespace *user_ns; /* Owning user namespace */ + unsigned int proc_inum; + struct proc_dir_entry *proc_net; struct proc_dir_entry *proc_net_stat; diff --git a/init/version.c b/init/version.c index 86fe0ccb997..58170f18912 100644 --- a/init/version.c +++ b/init/version.c @@ -12,6 +12,7 @@ #include #include #include +#include #ifndef CONFIG_KALLSYMS #define version(a) Version_ ## a @@ -34,6 +35,7 @@ struct uts_namespace init_uts_ns = { .domainname = UTS_DOMAINNAME, }, .user_ns = &init_user_ns, + .proc_inum = PROC_UTS_INIT_INO, }; EXPORT_SYMBOL_GPL(init_uts_ns); diff --git a/ipc/msgutil.c b/ipc/msgutil.c index 26143d377c9..6471f1bdae9 100644 --- a/ipc/msgutil.c +++ b/ipc/msgutil.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include "util.h" @@ -30,6 +31,7 @@ DEFINE_SPINLOCK(mq_lock); struct ipc_namespace init_ipc_ns = { .count = ATOMIC_INIT(1), .user_ns = &init_user_ns, + .proc_inum = PROC_IPC_INIT_INO, }; atomic_t nr_ipc_ns = ATOMIC_INIT(1); diff --git a/ipc/namespace.c b/ipc/namespace.c index 72c86827779..cf3386a51de 100644 --- a/ipc/namespace.c +++ b/ipc/namespace.c @@ -26,9 +26,16 @@ static struct ipc_namespace *create_ipc_ns(struct user_namespace *user_ns, if (ns == NULL) return ERR_PTR(-ENOMEM); + err = proc_alloc_inum(&ns->proc_inum); + if (err) { + kfree(ns); + return ERR_PTR(err); + } + atomic_set(&ns->count, 1); err = mq_init_ns(ns); if (err) { + proc_free_inum(ns->proc_inum); kfree(ns); return ERR_PTR(err); } @@ -111,6 +118,7 @@ static void free_ipc_ns(struct ipc_namespace *ns) */ ipcns_notify(IPCNS_REMOVED); put_user_ns(ns->user_ns); + proc_free_inum(ns->proc_inum); kfree(ns); } @@ -172,10 +180,18 @@ static int ipcns_install(struct nsproxy *nsproxy, void *new) return 0; } +static unsigned int ipcns_inum(void *vp) +{ + struct ipc_namespace *ns = vp; + + return ns->proc_inum; +} + const struct proc_ns_operations ipcns_operations = { .name = "ipc", .type = CLONE_NEWIPC, .get = ipcns_get, .put = ipcns_put, .install = ipcns_install, + .inum = ipcns_inum, }; diff --git a/kernel/pid.c b/kernel/pid.c index 6e8da291de4..3026ddae0a3 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -80,6 +80,7 @@ struct pid_namespace init_pid_ns = { .level = 0, .child_reaper = &init_task, .user_ns = &init_user_ns, + .proc_inum = PROC_PID_INIT_INO, }; EXPORT_SYMBOL_GPL(init_pid_ns); diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index 68508d33063..560da0dab23 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -107,6 +107,10 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns if (ns->pid_cachep == NULL) goto out_free_map; + err = proc_alloc_inum(&ns->proc_inum); + if (err) + goto out_free_map; + kref_init(&ns->kref); ns->level = level; ns->parent = get_pid_ns(parent_pid_ns); @@ -133,6 +137,7 @@ static void destroy_pid_namespace(struct pid_namespace *ns) { int i; + proc_free_inum(ns->proc_inum); for (i = 0; i < PIDMAP_ENTRIES; i++) kfree(ns->pidmap[i].page); put_user_ns(ns->user_ns); @@ -345,12 +350,19 @@ static int pidns_install(struct nsproxy *nsproxy, void *ns) return 0; } +static unsigned int pidns_inum(void *ns) +{ + struct pid_namespace *pid_ns = ns; + return pid_ns->proc_inum; +} + const struct proc_ns_operations pidns_operations = { .name = "pid", .type = CLONE_NEWPID, .get = pidns_get, .put = pidns_put, .install = pidns_install, + .inum = pidns_inum, }; static __init int pid_namespaces_init(void) diff --git a/kernel/user.c b/kernel/user.c index 750acffbe9e..33acb5e53a5 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -16,6 +16,7 @@ #include #include #include +#include /* * userns count is 1 for root user, 1 for init_uts_ns, @@ -51,6 +52,7 @@ struct user_namespace init_user_ns = { }, .owner = GLOBAL_ROOT_UID, .group = GLOBAL_ROOT_GID, + .proc_inum = PROC_USER_INIT_INO, }; EXPORT_SYMBOL_GPL(init_user_ns); diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index 89f6eaed067..f5975ccf934 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -58,6 +58,7 @@ int create_user_ns(struct cred *new) struct user_namespace *ns, *parent_ns = new->user_ns; kuid_t owner = new->euid; kgid_t group = new->egid; + int ret; /* The creator needs a mapping in the parent user namespace * or else we won't be able to reasonably tell userspace who @@ -71,6 +72,12 @@ int create_user_ns(struct cred *new) if (!ns) return -ENOMEM; + ret = proc_alloc_inum(&ns->proc_inum); + if (ret) { + kmem_cache_free(user_ns_cachep, ns); + return ret; + } + kref_init(&ns->kref); /* Leave the new->user_ns reference with the new user namespace. */ ns->parent = parent_ns; @@ -103,6 +110,7 @@ void free_user_ns(struct kref *kref) container_of(kref, struct user_namespace, kref); parent = ns->parent; + proc_free_inum(ns->proc_inum); kmem_cache_free(user_ns_cachep, ns); put_user_ns(parent); } @@ -808,12 +816,19 @@ static int userns_install(struct nsproxy *nsproxy, void *ns) return commit_creds(cred); } +static unsigned int userns_inum(void *ns) +{ + struct user_namespace *user_ns = ns; + return user_ns->proc_inum; +} + const struct proc_ns_operations userns_operations = { .name = "user", .type = CLONE_NEWUSER, .get = userns_get, .put = userns_put, .install = userns_install, + .inum = userns_inum, }; static __init int user_namespaces_init(void) diff --git a/kernel/utsname.c b/kernel/utsname.c index fdc619eb61e..f6336d51d64 100644 --- a/kernel/utsname.c +++ b/kernel/utsname.c @@ -36,11 +36,18 @@ static struct uts_namespace *clone_uts_ns(struct user_namespace *user_ns, struct uts_namespace *old_ns) { struct uts_namespace *ns; + int err; ns = create_uts_ns(); if (!ns) return ERR_PTR(-ENOMEM); + err = proc_alloc_inum(&ns->proc_inum); + if (err) { + kfree(ns); + return ERR_PTR(err); + } + down_read(&uts_sem); memcpy(&ns->name, &old_ns->name, sizeof(ns->name)); ns->user_ns = get_user_ns(user_ns); @@ -77,6 +84,7 @@ void free_uts_ns(struct kref *kref) ns = container_of(kref, struct uts_namespace, kref); put_user_ns(ns->user_ns); + proc_free_inum(ns->proc_inum); kfree(ns); } @@ -114,11 +122,18 @@ static int utsns_install(struct nsproxy *nsproxy, void *new) return 0; } +static unsigned int utsns_inum(void *vp) +{ + struct uts_namespace *ns = vp; + + return ns->proc_inum; +} + const struct proc_ns_operations utsns_operations = { .name = "uts", .type = CLONE_NEWUTS, .get = utsns_get, .put = utsns_put, .install = utsns_install, + .inum = utsns_inum, }; - diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index ec2870b44c1..2e9a3132b8d 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -381,6 +381,21 @@ struct net *get_net_ns_by_pid(pid_t pid) } EXPORT_SYMBOL_GPL(get_net_ns_by_pid); +static __net_init int net_ns_net_init(struct net *net) +{ + return proc_alloc_inum(&net->proc_inum); +} + +static __net_exit void net_ns_net_exit(struct net *net) +{ + proc_free_inum(net->proc_inum); +} + +static struct pernet_operations __net_initdata net_ns_ops = { + .init = net_ns_net_init, + .exit = net_ns_net_exit, +}; + static int __init net_ns_init(void) { struct net_generic *ng; @@ -412,6 +427,8 @@ static int __init net_ns_init(void) mutex_unlock(&net_mutex); + register_pernet_subsys(&net_ns_ops); + return 0; } @@ -640,11 +657,18 @@ static int netns_install(struct nsproxy *nsproxy, void *ns) return 0; } +static unsigned int netns_inum(void *ns) +{ + struct net *net = ns; + return net->proc_inum; +} + const struct proc_ns_operations netns_operations = { .name = "net", .type = CLONE_NEWNET, .get = netns_get, .put = netns_put, .install = netns_install, + .inum = netns_inum, }; #endif -- cgit v1.2.3-70-g09d2 From c876ad7682155958d0c9c27afe9017925c230d64 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Fri, 21 Dec 2012 20:27:12 -0800 Subject: pidns: Stop pid allocation when init dies Oleg pointed out that in a pid namespace the sequence. - pid 1 becomes a zombie - setns(thepidns), fork,... - reaping pid 1. - The injected processes exiting. Can lead to processes attempting access their child reaper and instead following a stale pointer. That waitpid for init can return before all of the processes in the pid namespace have exited is also unfortunate. Avoid these problems by disabling the allocation of new pids in a pid namespace when init dies, instead of when the last process in a pid namespace is reaped. Pointed-out-by: Oleg Nesterov Reviewed-by: Oleg Nesterov Signed-off-by: "Eric W. Biederman" --- include/linux/pid.h | 1 + include/linux/pid_namespace.h | 4 +++- kernel/pid.c | 15 ++++++++++++--- kernel/pid_namespace.c | 4 ++++ 4 files changed, 20 insertions(+), 4 deletions(-) (limited to 'include/linux/pid_namespace.h') diff --git a/include/linux/pid.h b/include/linux/pid.h index b152d44fb18..2381c973d89 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -121,6 +121,7 @@ int next_pidmap(struct pid_namespace *pid_ns, unsigned int last); extern struct pid *alloc_pid(struct pid_namespace *ns); extern void free_pid(struct pid *pid); +extern void disable_pid_allocation(struct pid_namespace *ns); /* * ns_of_pid() returns the pid namespace in which the specified pid was diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index bf285999273..215e5e3dda1 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -21,7 +21,7 @@ struct pid_namespace { struct kref kref; struct pidmap pidmap[PIDMAP_ENTRIES]; int last_pid; - int nr_hashed; + unsigned int nr_hashed; struct task_struct *child_reaper; struct kmem_cache *pid_cachep; unsigned int level; @@ -42,6 +42,8 @@ struct pid_namespace { extern struct pid_namespace init_pid_ns; +#define PIDNS_HASH_ADDING (1U << 31) + #ifdef CONFIG_PID_NS static inline struct pid_namespace *get_pid_ns(struct pid_namespace *ns) { diff --git a/kernel/pid.c b/kernel/pid.c index 36aa02ff17d..de9af600006 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -270,7 +270,6 @@ void free_pid(struct pid *pid) wake_up_process(ns->child_reaper); break; case 0: - ns->nr_hashed = -1; schedule_work(&ns->proc_work); break; } @@ -319,7 +318,7 @@ struct pid *alloc_pid(struct pid_namespace *ns) upid = pid->numbers + ns->level; spin_lock_irq(&pidmap_lock); - if (ns->nr_hashed < 0) + if (!(ns->nr_hashed & PIDNS_HASH_ADDING)) goto out_unlock; for ( ; upid >= pid->numbers; --upid) { hlist_add_head_rcu(&upid->pid_chain, @@ -342,6 +341,13 @@ out_free: goto out; } +void disable_pid_allocation(struct pid_namespace *ns) +{ + spin_lock_irq(&pidmap_lock); + ns->nr_hashed &= ~PIDNS_HASH_ADDING; + spin_unlock_irq(&pidmap_lock); +} + struct pid *find_pid_ns(int nr, struct pid_namespace *ns) { struct hlist_node *elem; @@ -573,6 +579,9 @@ void __init pidhash_init(void) void __init pidmap_init(void) { + /* Veryify no one has done anything silly */ + BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_HASH_ADDING); + /* bump default and minimum pid_max based on number of cpus */ pid_max = min(pid_max_max, max_t(int, pid_max, PIDS_PER_CPU_DEFAULT * num_possible_cpus())); @@ -584,7 +593,7 @@ void __init pidmap_init(void) /* Reserve PID 0. We never call free_pidmap(0) */ set_bit(0, init_pid_ns.pidmap[0].page); atomic_dec(&init_pid_ns.pidmap[0].nr_free); - init_pid_ns.nr_hashed = 1; + init_pid_ns.nr_hashed = PIDNS_HASH_ADDING; init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC); diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index fdbd0cdf271..c1c3dc1c602 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -115,6 +115,7 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns ns->level = level; ns->parent = get_pid_ns(parent_pid_ns); ns->user_ns = get_user_ns(user_ns); + ns->nr_hashed = PIDNS_HASH_ADDING; INIT_WORK(&ns->proc_work, proc_cleanup_work); set_bit(0, ns->pidmap[0].page); @@ -181,6 +182,9 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) int rc; struct task_struct *task, *me = current; + /* Don't allow any more processes into the pid namespace */ + disable_pid_allocation(pid_ns); + /* Ignore SIGCHLD causing any terminated children to autoreap */ spin_lock_irq(&me->sighand->siglock); me->sighand->action[SIGCHLD - 1].sa.sa_handler = SIG_IGN; -- cgit v1.2.3-70-g09d2