Merge branch 'android-4.4@e4528dd' into branch 'msm-4.4'

* refs/heads/tmp-e4528dd:
  Linux 4.4.65
  perf/core: Fix concurrent sys_perf_event_open() vs. 'move_group' race
  ping: implement proper locking
  staging/android/ion : fix a race condition in the ion driver
  vfio/pci: Fix integer overflows, bitmask check
  tipc: check minimum bearer MTU
  netfilter: nfnetlink: correctly validate length of batch messages
  xc2028: avoid use after free
  mnt: Add a per mount namespace limit on the number of mounts
  tipc: fix socket timer deadlock
  tipc: fix random link resets while adding a second bearer
  gfs2: avoid uninitialized variable warning
  hostap: avoid uninitialized variable use in hfa384x_get_rid
  tty: nozomi: avoid a harmless gcc warning
  tipc: correct error in node fsm
  tipc: re-enable compensation for socket receive buffer double counting
  tipc: make dist queue pernet
  tipc: make sure IPv6 header fits in skb headroom
  ANDROID: uid_sys_stats: fix access of task_uid(task)
  BACKPORT: f2fs: sanity check log_blocks_per_seg
  Linux 4.4.64
  tipc: fix crash during node removal
  block: fix del_gendisk() vs blkdev_ioctl crash
  x86, pmem: fix broken __copy_user_nocache cache-bypass assumptions
  hv: don't reset hv_context.tsc_page on crash
  Drivers: hv: balloon: account for gaps in hot add regions
  Drivers: hv: balloon: keep track of where ha_region starts
  Tools: hv: kvp: ensure kvp device fd is closed on exec
  kvm: arm/arm64: Fix locking for kvm_free_stage2_pgd
  x86/mce/AMD: Give a name to MCA bank 3 when accessed with legacy MSRs
  powerpc/kprobe: Fix oops when kprobed on 'stdu' instruction
  ubi/upd: Always flush after prepared for an update
  mac80211: reject ToDS broadcast data frames
  mmc: sdhci-esdhc-imx: increase the pad I/O drive strength for DDR50 card
  ACPI / power: Avoid maybe-uninitialized warning
  Input: elantech - add Fujitsu Lifebook E547 to force crc_enabled
  VSOCK: Detach QP check should filter out non matching QPs.
  Drivers: hv: vmbus: Reduce the delay between retries in vmbus_post_msg()
  Drivers: hv: get rid of timeout in vmbus_open()
  Drivers: hv: don't leak memory in vmbus_establish_gpadl()
  s390/mm: fix CMMA vs KSM vs others
  CIFS: remove bad_network_name flag
  cifs: Do not send echoes before Negotiate is complete
  ring-buffer: Have ring_buffer_iter_empty() return true when empty
  tracing: Allocate the snapshot buffer before enabling probe
  KEYS: fix keyctl_set_reqkey_keyring() to not leak thread keyrings
  KEYS: Change the name of the dead type to ".dead" to prevent user access
  KEYS: Disallow keyrings beginning with '.' to be joined as session keyrings
  ANDROID: sdcardfs: Call lower fs's revalidate
  ANDROID: sdcardfs: Avoid setting GIDs outside of valid ranges
  ANDROID: sdcardfs: Copy meta-data from lower inode
  Revert "Revert "Android: sdcardfs: Don't do d_add for lower fs""
  ANDROID: sdcardfs: Use filesystem specific hash
  ANDROID: AVB error handler to invalidate vbmeta partition.
  ANDROID: Update init/do_mounts_dm.c to the latest ChromiumOS version.
  Revert "[RFC]cgroup: Change from CAP_SYS_NICE to CAP_SYS_RESOURCE for cgroup migration permissions"

Conflicts:
	drivers/md/Makefile

Change-Id: I8f5ed53cb8b6cc66914f10c6ac820003b87b8759
Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>

1 files changed, 49 insertions, 1 deletions

diff --git a/fs/namespace.c b/fs/namespace.c
index c1477882a853..a22959c97384 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -27,6 +27,9 @@
 #include "pnode.h"
 #include "internal.h"
 
+/* Maximum number of mounts in a mount namespace */
+unsigned int sysctl_mount_max __read_mostly = 100000;
+
 static unsigned int m_hash_mask __read_mostly;
 static unsigned int m_hash_shift __read_mostly;
 static unsigned int mp_hash_mask __read_mostly;
@@ -926,6 +929,9 @@ static void commit_tree(struct mount *mnt)
 
 	list_splice(&head, n->list.prev);
 
+	n->mounts += n->pending_mounts;
+	n->pending_mounts = 0;
+
 	__attach_mnt(mnt, parent);
 	touch_mnt_namespace(n);
 }
@@ -1465,11 +1471,16 @@ static void umount_tree(struct mount *mnt, enum umount_tree_flags how)
 		propagate_umount(&tmp_list);
 
 	while (!list_empty(&tmp_list)) {
+		struct mnt_namespace *ns;
 		bool disconnect;
 		p = list_first_entry(&tmp_list, struct mount, mnt_list);
 		list_del_init(&p->mnt_expire);
 		list_del_init(&p->mnt_list);
-		__touch_mnt_namespace(p->mnt_ns);
+		ns = p->mnt_ns;
+		if (ns) {
+			ns->mounts--;
+			__touch_mnt_namespace(ns);
+		}
 		p->mnt_ns = NULL;
 		if (how & UMOUNT_SYNC)
 			p->mnt.mnt_flags |= MNT_SYNC_UMOUNT;
@@ -1870,6 +1881,28 @@ static int invent_group_ids(struct mount *mnt, bool recurse)
 	return 0;
 }
 
+int count_mounts(struct mnt_namespace *ns, struct mount *mnt)
+{
+	unsigned int max = READ_ONCE(sysctl_mount_max);
+	unsigned int mounts = 0, old, pending, sum;
+	struct mount *p;
+
+	for (p = mnt; p; p = next_mnt(p, mnt))
+		mounts++;
+
+	old = ns->mounts;
+	pending = ns->pending_mounts;
+	sum = old + pending;
+	if ((old > sum) ||
+	    (pending > sum) ||
+	    (max < sum) ||
+	    (mounts > (max - sum)))
+		return -ENOSPC;
+
+	ns->pending_mounts = pending + mounts;
+	return 0;
+}
+
 /*
  *  @source_mnt : mount tree to be attached
  *  @nd         : place the mount tree @source_mnt is attached
@@ -1939,6 +1972,7 @@ static int attach_recursive_mnt(struct mount *source_mnt,
 			struct path *parent_path)
 {
 	HLIST_HEAD(tree_list);
+	struct mnt_namespace *ns = dest_mnt->mnt_ns;
 	struct mountpoint *smp;
 	struct mount *child, *p;
 	struct hlist_node *n;
@@ -1951,6 +1985,13 @@ static int attach_recursive_mnt(struct mount *source_mnt,
 	if (IS_ERR(smp))
 		return PTR_ERR(smp);
 
+	/* Is there space to add these mounts to the mount namespace? */
+	if (!parent_path) {
+		err = count_mounts(ns, source_mnt);
+		if (err)
+			goto out;
+	}
+
 	if (IS_MNT_SHARED(dest_mnt)) {
 		err = invent_group_ids(source_mnt, true);
 		if (err)
@@ -1990,11 +2031,14 @@ static int attach_recursive_mnt(struct mount *source_mnt,
  out_cleanup_ids:
 	while (!hlist_empty(&tree_list)) {
 		child = hlist_entry(tree_list.first, struct mount, mnt_hash);
+		child->mnt_parent->mnt_ns->pending_mounts = 0;
 		umount_tree(child, UMOUNT_SYNC);
 	}
 	unlock_mount_hash();
 	cleanup_group_ids(source_mnt, NULL);
  out:
+	ns->pending_mounts = 0;
+
 	read_seqlock_excl(&mount_lock);
 	put_mountpoint(smp);
 	read_sequnlock_excl(&mount_lock);
@@ -2830,6 +2874,8 @@ static struct mnt_namespace *alloc_mnt_ns(struct user_namespace *user_ns)
 	init_waitqueue_head(&new_ns->poll);
 	new_ns->event = 0;
 	new_ns->user_ns = get_user_ns(user_ns);
+	new_ns->mounts = 0;
+	new_ns->pending_mounts = 0;
 	return new_ns;
 }
 
@@ -2879,6 +2925,7 @@ struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns,
 	q = new;
 	while (p) {
 		q->mnt_ns = new_ns;
+		new_ns->mounts++;
 		if (new_fs) {
 			if (&p->mnt == new_fs->root.mnt) {
 				new_fs->root.mnt = mntget(&q->mnt);
@@ -2917,6 +2964,7 @@ static struct mnt_namespace *create_mnt_ns(struct vfsmount *m)
 		struct mount *mnt = real_mount(m);
 		mnt->mnt_ns = new_ns;
 		new_ns->root = mnt;
+		new_ns->mounts++;
 		list_add(&mnt->mnt_list, &new_ns->list);
 	} else {
 		mntput(m);