summaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
authorJiri Kosina <jkosina@suse.cz>2010-08-11 09:36:51 +0200
committerJiri Kosina <jkosina@suse.cz>2010-08-11 09:36:51 +0200
commit6396fc3b3ff3f6b942992b653a62df11dcef9bea (patch)
treedb3c7cbe833b43c653adc99f70941431c5ff7c4e /kernel
parent4785879e4d340e24e54f6de2ccfc42728b912808 (diff)
parent3d30701b58970425e1d45994d6cb82f828924fdd (diff)
Merge branch 'master' into for-next
Conflicts: fs/exofs/inode.c
Diffstat (limited to 'kernel')
-rw-r--r--kernel/Makefile5
-rw-r--r--kernel/acct.c2
-rw-r--r--kernel/audit.c1
-rw-r--r--kernel/audit.h26
-rw-r--r--kernel/audit_tree.c237
-rw-r--r--kernel/audit_watch.c274
-rw-r--r--kernel/auditfilter.c39
-rw-r--r--kernel/auditsc.c10
-rw-r--r--kernel/compat.c17
-rw-r--r--kernel/posix-cpu-timers.c8
-rw-r--r--kernel/power/block_io.c2
-rw-r--r--kernel/sys.c202
-rw-r--r--kernel/sys_ni.c4
-rw-r--r--kernel/sysctl.c7
-rw-r--r--kernel/timer.c1
-rw-r--r--kernel/trace/blktrace.c80
16 files changed, 556 insertions, 359 deletions
diff --git a/kernel/Makefile b/kernel/Makefile
index c53e491e25a..0b72d1a74be 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -70,10 +70,11 @@ obj-$(CONFIG_IKCONFIG) += configs.o
obj-$(CONFIG_RESOURCE_COUNTERS) += res_counter.o
obj-$(CONFIG_SMP) += stop_machine.o
obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
-obj-$(CONFIG_AUDIT) += audit.o auditfilter.o audit_watch.o
+obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
-obj-$(CONFIG_GCOV_KERNEL) += gcov/
+obj-$(CONFIG_AUDIT_WATCH) += audit_watch.o
obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
+obj-$(CONFIG_GCOV_KERNEL) += gcov/
obj-$(CONFIG_KPROBES) += kprobes.o
obj-$(CONFIG_KGDB) += debug/
obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o
diff --git a/kernel/acct.c b/kernel/acct.c
index 385b88461c2..fa7eb3de2dd 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -122,7 +122,7 @@ static int check_free_space(struct bsd_acct_struct *acct, struct file *file)
spin_unlock(&acct_lock);
/* May block */
- if (vfs_statfs(file->f_path.dentry, &sbuf))
+ if (vfs_statfs(&file->f_path, &sbuf))
return res;
suspend = sbuf.f_blocks * SUSPEND;
resume = sbuf.f_blocks * RESUME;
diff --git a/kernel/audit.c b/kernel/audit.c
index 8296aa516c5..d96045789b5 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -56,7 +56,6 @@
#include <net/netlink.h>
#include <linux/skbuff.h>
#include <linux/netlink.h>
-#include <linux/inotify.h>
#include <linux/freezer.h>
#include <linux/tty.h>
diff --git a/kernel/audit.h b/kernel/audit.h
index 208687be4f3..f7206db4e13 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -103,21 +103,27 @@ extern struct mutex audit_filter_mutex;
extern void audit_free_rule_rcu(struct rcu_head *);
extern struct list_head audit_filter_list[];
+extern struct audit_entry *audit_dupe_rule(struct audit_krule *old);
+
/* audit watch functions */
-extern unsigned long audit_watch_inode(struct audit_watch *watch);
-extern dev_t audit_watch_dev(struct audit_watch *watch);
+#ifdef CONFIG_AUDIT_WATCH
extern void audit_put_watch(struct audit_watch *watch);
extern void audit_get_watch(struct audit_watch *watch);
extern int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op);
-extern int audit_add_watch(struct audit_krule *krule);
-extern void audit_remove_watch(struct audit_watch *watch);
-extern void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list);
-extern void audit_inotify_unregister(struct list_head *in_list);
+extern int audit_add_watch(struct audit_krule *krule, struct list_head **list);
+extern void audit_remove_watch_rule(struct audit_krule *krule);
extern char *audit_watch_path(struct audit_watch *watch);
-extern struct list_head *audit_watch_rules(struct audit_watch *watch);
-
-extern struct audit_entry *audit_dupe_rule(struct audit_krule *old,
- struct audit_watch *watch);
+extern int audit_watch_compare(struct audit_watch *watch, unsigned long ino, dev_t dev);
+#else
+#define audit_put_watch(w) {}
+#define audit_get_watch(w) {}
+#define audit_to_watch(k, p, l, o) (-EINVAL)
+#define audit_add_watch(k, l) (-EINVAL)
+#define audit_remove_watch_rule(k) BUG()
+#define audit_watch_path(w) ""
+#define audit_watch_compare(w, i, d) 0
+
+#endif /* CONFIG_AUDIT_WATCH */
#ifdef CONFIG_AUDIT_TREE
extern struct audit_chunk *audit_tree_lookup(const struct inode *);
diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c
index 46a57b57a33..7f18d3a4527 100644
--- a/kernel/audit_tree.c
+++ b/kernel/audit_tree.c
@@ -1,5 +1,5 @@
#include "audit.h"
-#include <linux/inotify.h>
+#include <linux/fsnotify_backend.h>
#include <linux/namei.h>
#include <linux/mount.h>
#include <linux/kthread.h>
@@ -22,7 +22,7 @@ struct audit_tree {
struct audit_chunk {
struct list_head hash;
- struct inotify_watch watch;
+ struct fsnotify_mark mark;
struct list_head trees; /* with root here */
int dead;
int count;
@@ -59,7 +59,7 @@ static LIST_HEAD(prune_list);
* tree is refcounted; one reference for "some rules on rules_list refer to
* it", one for each chunk with pointer to it.
*
- * chunk is refcounted by embedded inotify_watch + .refs (non-zero refcount
+ * chunk is refcounted by embedded fsnotify_mark + .refs (non-zero refcount
* of watch contributes 1 to .refs).
*
* node.index allows to get from node.list to containing chunk.
@@ -68,7 +68,7 @@ static LIST_HEAD(prune_list);
* that makes a difference. Some.
*/
-static struct inotify_handle *rtree_ih;
+static struct fsnotify_group *audit_tree_group;
static struct audit_tree *alloc_tree(const char *s)
{
@@ -111,29 +111,6 @@ const char *audit_tree_path(struct audit_tree *tree)
return tree->pathname;
}
-static struct audit_chunk *alloc_chunk(int count)
-{
- struct audit_chunk *chunk;
- size_t size;
- int i;
-
- size = offsetof(struct audit_chunk, owners) + count * sizeof(struct node);
- chunk = kzalloc(size, GFP_KERNEL);
- if (!chunk)
- return NULL;
-
- INIT_LIST_HEAD(&chunk->hash);
- INIT_LIST_HEAD(&chunk->trees);
- chunk->count = count;
- atomic_long_set(&chunk->refs, 1);
- for (i = 0; i < count; i++) {
- INIT_LIST_HEAD(&chunk->owners[i].list);
- chunk->owners[i].index = i;
- }
- inotify_init_watch(&chunk->watch);
- return chunk;
-}
-
static void free_chunk(struct audit_chunk *chunk)
{
int i;
@@ -157,6 +134,35 @@ static void __put_chunk(struct rcu_head *rcu)
audit_put_chunk(chunk);
}
+static void audit_tree_destroy_watch(struct fsnotify_mark *entry)
+{
+ struct audit_chunk *chunk = container_of(entry, struct audit_chunk, mark);
+ call_rcu(&chunk->head, __put_chunk);
+}
+
+static struct audit_chunk *alloc_chunk(int count)
+{
+ struct audit_chunk *chunk;
+ size_t size;
+ int i;
+
+ size = offsetof(struct audit_chunk, owners) + count * sizeof(struct node);
+ chunk = kzalloc(size, GFP_KERNEL);
+ if (!chunk)
+ return NULL;
+
+ INIT_LIST_HEAD(&chunk->hash);
+ INIT_LIST_HEAD(&chunk->trees);
+ chunk->count = count;
+ atomic_long_set(&chunk->refs, 1);
+ for (i = 0; i < count; i++) {
+ INIT_LIST_HEAD(&chunk->owners[i].list);
+ chunk->owners[i].index = i;
+ }
+ fsnotify_init_mark(&chunk->mark, audit_tree_destroy_watch);
+ return chunk;
+}
+
enum {HASH_SIZE = 128};
static struct list_head chunk_hash_heads[HASH_SIZE];
static __cacheline_aligned_in_smp DEFINE_SPINLOCK(hash_lock);
@@ -167,10 +173,15 @@ static inline struct list_head *chunk_hash(const struct inode *inode)
return chunk_hash_heads + n % HASH_SIZE;
}
-/* hash_lock is held by caller */
+/* hash_lock & entry->lock is held by caller */
static void insert_hash(struct audit_chunk *chunk)
{
- struct list_head *list = chunk_hash(chunk->watch.inode);
+ struct fsnotify_mark *entry = &chunk->mark;
+ struct list_head *list;
+
+ if (!entry->i.inode)
+ return;
+ list = chunk_hash(entry->i.inode);
list_add_rcu(&chunk->hash, list);
}
@@ -181,7 +192,8 @@ struct audit_chunk *audit_tree_lookup(const struct inode *inode)
struct audit_chunk *p;
list_for_each_entry_rcu(p, list, hash) {
- if (p->watch.inode == inode) {
+ /* mark.inode may have gone NULL, but who cares? */
+ if (p->mark.i.inode == inode) {
atomic_long_inc(&p->refs);
return p;
}
@@ -210,38 +222,19 @@ static struct audit_chunk *find_chunk(struct node *p)
static void untag_chunk(struct node *p)
{
struct audit_chunk *chunk = find_chunk(p);
+ struct fsnotify_mark *entry = &chunk->mark;
struct audit_chunk *new;
struct audit_tree *owner;
int size = chunk->count - 1;
int i, j;
- if (!pin_inotify_watch(&chunk->watch)) {
- /*
- * Filesystem is shutting down; all watches are getting
- * evicted, just take it off the node list for this
- * tree and let the eviction logics take care of the
- * rest.
- */
- owner = p->owner;
- if (owner->root == chunk) {
- list_del_init(&owner->same_root);
- owner->root = NULL;
- }
- list_del_init(&p->list);
- p->owner = NULL;
- put_tree(owner);
- return;
- }
+ fsnotify_get_mark(entry);
spin_unlock(&hash_lock);
- /*
- * pin_inotify_watch() succeeded, so the watch won't go away
- * from under us.
- */
- mutex_lock(&chunk->watch.inode->inotify_mutex);
- if (chunk->dead) {
- mutex_unlock(&chunk->watch.inode->inotify_mutex);
+ spin_lock(&entry->lock);
+ if (chunk->dead || !entry->i.inode) {
+ spin_unlock(&entry->lock);
goto out;
}
@@ -256,16 +249,17 @@ static void untag_chunk(struct node *p)
list_del_init(&p->list);
list_del_rcu(&chunk->hash);
spin_unlock(&hash_lock);
- inotify_evict_watch(&chunk->watch);
- mutex_unlock(&chunk->watch.inode->inotify_mutex);
- put_inotify_watch(&chunk->watch);
+ spin_unlock(&entry->lock);
+ fsnotify_destroy_mark(entry);
+ fsnotify_put_mark(entry);
goto out;
}
new = alloc_chunk(size);
if (!new)
goto Fallback;
- if (inotify_clone_watch(&chunk->watch, &new->watch) < 0) {
+ fsnotify_duplicate_mark(&new->mark, entry);
+ if (fsnotify_add_mark(&new->mark, new->mark.group, new->mark.i.inode, NULL, 1)) {
free_chunk(new);
goto Fallback;
}
@@ -298,9 +292,9 @@ static void untag_chunk(struct node *p)
list_for_each_entry(owner, &new->trees, same_root)
owner->root = new;
spin_unlock(&hash_lock);
- inotify_evict_watch(&chunk->watch);
- mutex_unlock(&chunk->watch.inode->inotify_mutex);
- put_inotify_watch(&chunk->watch);
+ spin_unlock(&entry->lock);
+ fsnotify_destroy_mark(entry);
+ fsnotify_put_mark(entry);
goto out;
Fallback:
@@ -314,31 +308,33 @@ Fallback:
p->owner = NULL;
put_tree(owner);
spin_unlock(&hash_lock);
- mutex_unlock(&chunk->watch.inode->inotify_mutex);
+ spin_unlock(&entry->lock);
out:
- unpin_inotify_watch(&chunk->watch);
+ fsnotify_put_mark(entry);
spin_lock(&hash_lock);
}
static int create_chunk(struct inode *inode, struct audit_tree *tree)
{
+ struct fsnotify_mark *entry;
struct audit_chunk *chunk = alloc_chunk(1);
if (!chunk)
return -ENOMEM;
- if (inotify_add_watch(rtree_ih, &chunk->watch, inode, IN_IGNORED | IN_DELETE_SELF) < 0) {
+ entry = &chunk->mark;
+ if (fsnotify_add_mark(entry, audit_tree_group, inode, NULL, 0)) {
free_chunk(chunk);
return -ENOSPC;
}
- mutex_lock(&inode->inotify_mutex);
+ spin_lock(&entry->lock);
spin_lock(&hash_lock);
if (tree->goner) {
spin_unlock(&hash_lock);
chunk->dead = 1;
- inotify_evict_watch(&chunk->watch);
- mutex_unlock(&inode->inotify_mutex);
- put_inotify_watch(&chunk->watch);
+ spin_unlock(&entry->lock);
+ fsnotify_destroy_mark(entry);
+ fsnotify_put_mark(entry);
return 0;
}
chunk->owners[0].index = (1U << 31);
@@ -351,30 +347,31 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree)
}
insert_hash(chunk);
spin_unlock(&hash_lock);
- mutex_unlock(&inode->inotify_mutex);
+ spin_unlock(&entry->lock);
return 0;
}
/* the first tagged inode becomes root of tree */
static int tag_chunk(struct inode *inode, struct audit_tree *tree)
{
- struct inotify_watch *watch;
+ struct fsnotify_mark *old_entry, *chunk_entry;
struct audit_tree *owner;
struct audit_chunk *chunk, *old;
struct node *p;
int n;
- if (inotify_find_watch(rtree_ih, inode, &watch) < 0)
+ old_entry = fsnotify_find_inode_mark(audit_tree_group, inode);
+ if (!old_entry)
return create_chunk(inode, tree);
- old = container_of(watch, struct audit_chunk, watch);
+ old = container_of(old_entry, struct audit_chunk, mark);
/* are we already there? */
spin_lock(&hash_lock);
for (n = 0; n < old->count; n++) {
if (old->owners[n].owner == tree) {
spin_unlock(&hash_lock);
- put_inotify_watch(&old->watch);
+ fsnotify_put_mark(old_entry);
return 0;
}
}
@@ -382,25 +379,44 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree)
chunk = alloc_chunk(old->count + 1);
if (!chunk) {
- put_inotify_watch(&old->watch);
+ fsnotify_put_mark(old_entry);
return -ENOMEM;
}
- mutex_lock(&inode->inotify_mutex);
- if (inotify_clone_watch(&old->watch, &chunk->watch) < 0) {
- mutex_unlock(&inode->inotify_mutex);
- put_inotify_watch(&old->watch);
+ chunk_entry = &chunk->mark;
+
+ spin_lock(&old_entry->lock);
+ if (!old_entry->i.inode) {
+ /* old_entry is being shot, lets just lie */
+ spin_unlock(&old_entry->lock);
+ fsnotify_put_mark(old_entry);
free_chunk(chunk);
+ return -ENOENT;
+ }
+
+ fsnotify_duplicate_mark(chunk_entry, old_entry);
+ if (fsnotify_add_mark(chunk_entry, chunk_entry->group, chunk_entry->i.inode, NULL, 1)) {
+ spin_unlock(&old_entry->lock);
+ free_chunk(chunk);
+ fsnotify_put_mark(old_entry);
return -ENOSPC;
}
+
+ /* even though we hold old_entry->lock, this is safe since chunk_entry->lock could NEVER have been grabbed before */
+ spin_lock(&chunk_entry->lock);
spin_lock(&hash_lock);
+
+ /* we now hold old_entry->lock, chunk_entry->lock, and hash_lock */
if (tree->goner) {
spin_unlock(&hash_lock);
chunk->dead = 1;
- inotify_evict_watch(&chunk->watch);
- mutex_unlock(&inode->inotify_mutex);
- put_inotify_watch(&old->watch);
- put_inotify_watch(&chunk->watch);
+ spin_unlock(&chunk_entry->lock);
+ spin_unlock(&old_entry->lock);
+
+ fsnotify_destroy_mark(chunk_entry);
+
+ fsnotify_put_mark(chunk_entry);
+ fsnotify_put_mark(old_entry);
return 0;
}
list_replace_init(&old->trees, &chunk->trees);
@@ -426,10 +442,11 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree)
list_add(&tree->same_root, &chunk->trees);
}
spin_unlock(&hash_lock);
- inotify_evict_watch(&old->watch);
- mutex_unlock(&inode->inotify_mutex);
- put_inotify_watch(&old->watch); /* pair to inotify_find_watch */
- put_inotify_watch(&old->watch); /* and kill it */
+ spin_unlock(&chunk_entry->lock);
+ spin_unlock(&old_entry->lock);
+ fsnotify_destroy_mark(old_entry);
+ fsnotify_put_mark(old_entry); /* pair to fsnotify_find mark_entry */
+ fsnotify_put_mark(old_entry); /* and kill it */
return 0;
}
@@ -584,7 +601,9 @@ void audit_trim_trees(void)
spin_lock(&hash_lock);
list_for_each_entry(node, &tree->chunks, list) {
- struct inode *inode = find_chunk(node)->watch.inode;
+ struct audit_chunk *chunk = find_chunk(node);
+ /* this could be NULL if the watch is dieing else where... */
+ struct inode *inode = chunk->mark.i.inode;
node->index |= 1U<<31;
if (iterate_mounts(compare_root, inode, root_mnt))
node->index &= ~(1U<<31);
@@ -846,7 +865,6 @@ void audit_kill_trees(struct list_head *list)
* Here comes the stuff asynchronous to auditctl operations
*/
-/* inode->inotify_mutex is locked */
static void evict_chunk(struct audit_chunk *chunk)
{
struct audit_tree *owner;
@@ -885,35 +903,46 @@ static void evict_chunk(struct audit_chunk *chunk)
mutex_unlock(&audit_filter_mutex);
}
-static void handle_event(struct inotify_watch *watch, u32 wd, u32 mask,
- u32 cookie, const char *dname, struct inode *inode)
+static int audit_tree_handle_event(struct fsnotify_group *group,
+ struct fsnotify_mark *inode_mark,
+ struct fsnotify_mark *vfsmonut_mark,
+ struct fsnotify_event *event)
+{
+ BUG();
+ return -EOPNOTSUPP;
+}
+
+static void audit_tree_freeing_mark(struct fsnotify_mark *entry, struct fsnotify_group *group)
{
- struct audit_chunk *chunk = container_of(watch, struct audit_chunk, watch);
+ struct audit_chunk *chunk = container_of(entry, struct audit_chunk, mark);
- if (mask & IN_IGNORED) {
- evict_chunk(chunk);
- put_inotify_watch(watch);
- }
+ evict_chunk(chunk);
+ fsnotify_put_mark(entry);
}
-static void destroy_watch(struct inotify_watch *watch)
+static bool audit_tree_send_event(struct fsnotify_group *group, struct inode *inode,
+ struct fsnotify_mark *inode_mark,
+ struct fsnotify_mark *vfsmount_mark,
+ __u32 mask, void *data, int data_type)
{
- struct audit_chunk *chunk = container_of(watch, struct audit_chunk, watch);
- call_rcu(&chunk->head, __put_chunk);
+ return false;
}
-static const struct inotify_operations rtree_inotify_ops = {
- .handle_event = handle_event,
- .destroy_watch = destroy_watch,
+static const struct fsnotify_ops audit_tree_ops = {
+ .handle_event = audit_tree_handle_event,
+ .should_send_event = audit_tree_send_event,
+ .free_group_priv = NULL,
+ .free_event_priv = NULL,
+ .freeing_mark = audit_tree_freeing_mark,
};
static int __init audit_tree_init(void)
{
int i;
- rtree_ih = inotify_init(&rtree_inotify_ops);
- if (IS_ERR(rtree_ih))
- audit_panic("cannot initialize inotify handle for rectree watches");
+ audit_tree_group = fsnotify_alloc_group(&audit_tree_ops);
+ if (IS_ERR(audit_tree_group))
+ audit_panic("cannot initialize fsnotify group for rectree watches");
for (i = 0; i < HASH_SIZE; i++)
INIT_LIST_HEAD(&chunk_hash_heads[i]);
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index 8df43696f4b..6bf2306be7d 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -24,18 +24,18 @@
#include <linux/kthread.h>
#include <linux/mutex.h>
#include <linux/fs.h>
+#include <linux/fsnotify_backend.h>
#include <linux/namei.h>
#include <linux/netlink.h>
#include <linux/sched.h>
#include <linux/slab.h>
-#include <linux/inotify.h>
#include <linux/security.h>
#include "audit.h"
/*
* Reference counting:
*
- * audit_parent: lifetime is from audit_init_parent() to receipt of an IN_IGNORED
+ * audit_parent: lifetime is from audit_init_parent() to receipt of an FS_IGNORED
* event. Each audit_watch holds a reference to its associated parent.
*
* audit_watch: if added to lists, lifetime is from audit_init_watch() to
@@ -51,40 +51,61 @@ struct audit_watch {
unsigned long ino; /* associated inode number */
struct audit_parent *parent; /* associated parent */
struct list_head wlist; /* entry in parent->watches list */
- struct list_head rules; /* associated rules */
+ struct list_head rules; /* anchor for krule->rlist */
};
struct audit_parent {
- struct list_head ilist; /* entry in inotify registration list */
- struct list_head watches; /* associated watches */
- struct inotify_watch wdata; /* inotify watch data */
- unsigned flags; /* status flags */
+ struct list_head watches; /* anchor for audit_watch->wlist */
+ struct fsnotify_mark mark; /* fsnotify mark on the inode */
};
-/* Inotify handle. */
-struct inotify_handle *audit_ih;
+/* fsnotify handle. */
+struct fsnotify_group *audit_watch_group;
-/*
- * audit_parent status flags:
- *
- * AUDIT_PARENT_INVALID - set anytime rules/watches are auto-removed due to
- * a filesystem event to ensure we're adding audit watches to a valid parent.
- * Technically not needed for IN_DELETE_SELF or IN_UNMOUNT events, as we cannot
- * receive them while we have nameidata, but must be used for IN_MOVE_SELF which
- * we can receive while holding nameidata.
- */
-#define AUDIT_PARENT_INVALID 0x001
+/* fsnotify events we care about. */
+#define AUDIT_FS_WATCH (FS_MOVE | FS_CREATE | FS_DELETE | FS_DELETE_SELF |\
+ FS_MOVE_SELF | FS_EVENT_ON_CHILD)
-/* Inotify events we care about. */
-#define AUDIT_IN_WATCH IN_MOVE|IN_CREATE|IN_DELETE|IN_DELETE_SELF|IN_MOVE_SELF
+static void audit_free_parent(struct audit_parent *parent)
+{
+ WARN_ON(!list_empty(&parent->watches));
+ kfree(parent);
+}
-static void audit_free_parent(struct inotify_watch *i_watch)
+static void audit_watch_free_mark(struct fsnotify_mark *entry)
{
struct audit_parent *parent;
- parent = container_of(i_watch, struct audit_parent, wdata);
- WARN_ON(!list_empty(&parent->watches));
- kfree(parent);
+ parent = container_of(entry, struct audit_parent, mark);
+ audit_free_parent(parent);
+}
+
+static void audit_get_parent(struct audit_parent *parent)
+{
+ if (likely(parent))
+ fsnotify_get_mark(&parent->mark);
+}
+
+static void audit_put_parent(struct audit_parent *parent)
+{
+ if (likely(parent))
+ fsnotify_put_mark(&parent->mark);
+}
+
+/*
+ * Find and return the audit_parent on the given inode. If found a reference
+ * is taken on this parent.
+ */
+static inline struct audit_parent *audit_find_parent(struct inode *inode)
+{
+ struct audit_parent *parent = NULL;
+ struct fsnotify_mark *entry;
+
+ entry = fsnotify_find_inode_mark(audit_watch_group, inode);
+ if (entry)
+ parent = container_of(entry, struct audit_parent, mark);
+
+ return parent;
}
void audit_get_watch(struct audit_watch *watch)
@@ -105,7 +126,7 @@ void audit_put_watch(struct audit_watch *watch)
void audit_remove_watch(struct audit_watch *watch)
{
list_del(&watch->wlist);
- put_inotify_watch(&watch->parent->wdata);
+ audit_put_parent(watch->parent);
watch->parent = NULL;
audit_put_watch(watch); /* match initial get */
}
@@ -115,42 +136,32 @@ char *audit_watch_path(struct audit_watch *watch)
return watch->path;
}
-struct list_head *audit_watch_rules(struct audit_watch *watch)
-{
- return &watch->rules;
-}
-
-unsigned long audit_watch_inode(struct audit_watch *watch)
+int audit_watch_compare(struct audit_watch *watch, unsigned long ino, dev_t dev)
{
- return watch->ino;
-}
-
-dev_t audit_watch_dev(struct audit_watch *watch)
-{
- return watch->dev;
+ return (watch->ino != (unsigned long)-1) &&
+ (watch->ino == ino) &&
+ (watch->dev == dev);
}
/* Initialize a parent watch entry. */
static struct audit_parent *audit_init_parent(struct nameidata *ndp)
{
+ struct inode *inode = ndp->path.dentry->d_inode;
struct audit_parent *parent;
- s32 wd;
+ int ret;
parent = kzalloc(sizeof(*parent), GFP_KERNEL);
if (unlikely(!parent))
return ERR_PTR(-ENOMEM);
INIT_LIST_HEAD(&parent->watches);
- parent->flags = 0;
-
- inotify_init_watch(&parent->wdata);
- /* grab a ref so inotify watch hangs around until we take audit_filter_mutex */
- get_inotify_watch(&parent->wdata);
- wd = inotify_add_watch(audit_ih, &parent->wdata,
- ndp->path.dentry->d_inode, AUDIT_IN_WATCH);
- if (wd < 0) {
- audit_free_parent(&parent->wdata);
- return ERR_PTR(wd);
+
+ fsnotify_init_mark(&parent->mark, audit_watch_free_mark);
+ parent->mark.mask = AUDIT_FS_WATCH;
+ ret = fsnotify_add_mark(&parent->mark, audit_watch_group, inode, NULL, 0);
+ if (ret < 0) {
+ audit_free_parent(parent);
+ return ERR_PTR(ret);
}
return parent;
@@ -179,7 +190,7 @@ int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op)
{
struct audit_watch *watch;
- if (!audit_ih)
+ if (!audit_watch_group)
return -EOPNOTSUPP;
if (path[0] != '/' || path[len-1] == '/' ||
@@ -217,7 +228,7 @@ static struct audit_watch *audit_dupe_watch(struct audit_watch *old)
new->dev = old->dev;
new->ino = old->ino;
- get_inotify_watch(&old->parent->wdata);
+ audit_get_parent(old->parent);
new->parent = old->parent;
out:
@@ -251,15 +262,19 @@ static void audit_update_watch(struct audit_parent *parent,
struct audit_entry *oentry, *nentry;
mutex_lock(&audit_filter_mutex);
+ /* Run all of the watches on this parent looking for the one that
+ * matches the given dname */
list_for_each_entry_safe(owatch, nextw, &parent->watches, wlist) {
if (audit_compare_dname_path(dname, owatch->path, NULL))
continue;
/* If the update involves invalidating rules, do the inode-based
* filtering now, so we don't omit records. */
- if (invalidating && current->audit_context)
+ if (invalidating && !audit_dummy_context())
audit_filter_inodes(current, current->audit_context);
+ /* updating ino will likely change which audit_hash_list we
+ * are on so we need a new watch for the new list */
nwatch = audit_dupe_watch(owatch);
if (IS_ERR(nwatch)) {
mutex_unlock(&audit_filter_mutex);
@@ -275,12 +290,21 @@ static void audit_update_watch(struct audit_parent *parent,
list_del(&oentry->rule.rlist);
list_del_rcu(&oentry->list);
- nentry = audit_dupe_rule(&oentry->rule, nwatch);
+ nentry = audit_dupe_rule(&oentry->rule);
if (IS_ERR(nentry)) {
list_del(&oentry->rule.list);
audit_panic("error updating watch, removing");
} else {
int h = audit_hash_ino((u32)ino);
+
+ /*
+ * nentry->rule.watch == oentry->rule.watch so
+ * we must drop that reference and set it to our
+ * new watch.
+ */
+ audit_put_watch(nentry->rule.watch);
+ audit_get_watch(nwatch);
+ nentry->rule.watch = nwatch;
list_add(&nentry->rule.rlist, &nwatch->rules);
list_add_rcu(&nentry->list, &audit_inode_hash[h]);
list_replace(&oentry->rule.list,
@@ -312,7 +336,6 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
struct audit_entry *e;
mutex_lock(&audit_filter_mutex);
- parent->flags |= AUDIT_PARENT_INVALID;
list_for_each_entry_safe(w, nextw, &parent->watches, wlist) {
list_for_each_entry_safe(r, nextr, &w->rules, rlist) {
e = container_of(r, struct audit_entry, rule);
@@ -325,20 +348,8 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
audit_remove_watch(w);
}
mutex_unlock(&audit_filter_mutex);
-}
-
-/* Unregister inotify watches for parents on in_list.
- * Generates an IN_IGNORED event. */
-void audit_inotify_unregister(struct list_head *in_list)
-{
- struct audit_parent *p, *n;
- list_for_each_entry_safe(p, n, in_list, ilist) {
- list_del(&p->ilist);
- inotify_rm_watch(audit_ih, &p->wdata);
- /* the unpin matching the pin in audit_do_del_rule() */
- unpin_inotify_watch(&p->wdata);
- }
+ fsnotify_destroy_mark(&parent->mark);
}
/* Get path information necessary for adding watches. */
@@ -389,7 +400,7 @@ static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw)
}
}
-/* Associate the given rule with an existing parent inotify_watch.
+/* Associate the given rule with an existing parent.
* Caller must hold audit_filter_mutex. */
static void audit_add_to_parent(struct audit_krule *krule,
struct audit_parent *parent)
@@ -397,6 +408,8 @@ static void audit_add_to_parent(struct audit_krule *krule,
struct audit_watch *w, *watch = krule->watch;
int watch_found = 0;
+ BUG_ON(!mutex_is_locked(&audit_filter_mutex));
+
list_for_each_entry(w, &parent->watches, wlist) {
if (strcmp(watch->path, w->path))
continue;
@@ -413,7 +426,7 @@ static void audit_add_to_parent(struct audit_krule *krule,
}
if (!watch_found) {
- get_inotify_watch(&parent->wdata);
+ audit_get_parent(parent);
watch->parent = parent;
list_add(&watch->wlist, &parent->watches);
@@ -423,13 +436,12 @@ static void audit_add_to_parent(struct audit_krule *krule,
/* Find a matching watch entry, or add this one.
* Caller must hold audit_filter_mutex. */
-int audit_add_watch(struct audit_krule *krule)
+int audit_add_watch(struct audit_krule *krule, struct list_head **list)
{
struct audit_watch *watch = krule->watch;
- struct inotify_watch *i_watch;
struct audit_parent *parent;
struct nameidata *ndp = NULL, *ndw = NULL;
- int ret = 0;
+ int h, ret = 0;
mutex_unlock(&audit_filter_mutex);
@@ -441,47 +453,38 @@ int audit_add_watch(struct audit_krule *krule)
goto error;
}
+ mutex_lock(&audit_filter_mutex);
+
/* update watch filter fields */
if (ndw) {
watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev;
watch->ino = ndw->path.dentry->d_inode->i_ino;
}
- /* The audit_filter_mutex must not be held during inotify calls because
- * we hold it during inotify event callback processing. If an existing
- * inotify watch is found, inotify_find_watch() grabs a reference before
- * returning.
- */
- if (inotify_find_watch(audit_ih, ndp->path.dentry->d_inode,
- &i_watch) < 0) {
+ /* either find an old parent or attach a new one */
+ parent = audit_find_parent(ndp->path.dentry->d_inode);
+ if (!parent) {
parent = audit_init_parent(ndp);
if (IS_ERR(parent)) {
- /* caller expects mutex locked */
- mutex_lock(&audit_filter_mutex);
ret = PTR_ERR(parent);
goto error;
}
- } else
- parent = container_of(i_watch, struct audit_parent, wdata);
-
- mutex_lock(&audit_filter_mutex);
+ }
- /* parent was moved before we took audit_filter_mutex */
- if (parent->flags & AUDIT_PARENT_INVALID)
- ret = -ENOENT;
- else
- audit_add_to_parent(krule, parent);
+ audit_add_to_parent(krule, parent);
- /* match get in audit_init_parent or inotify_find_watch */
- put_inotify_watch(&parent->wdata);
+ /* match get in audit_find_parent or audit_init_parent */
+ audit_put_parent(parent);
+ h = audit_hash_ino((u32)watch->ino);
+ *list = &audit_inode_hash[h];
error:
audit_put_nd(ndp, ndw); /* NULL args OK */
return ret;
}
-void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list)
+void audit_remove_watch_rule(struct audit_krule *krule)
{
struct audit_watch *watch = krule->watch;
struct audit_parent *parent = watch->parent;
@@ -492,53 +495,74 @@ void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list)
audit_remove_watch(watch);
if (list_empty(&parent->watches)) {
- /* Put parent on the inotify un-registration
- * list. Grab a reference before releasing
- * audit_filter_mutex, to be released in
- * audit_inotify_unregister().
- * If filesystem is going away, just leave
- * the sucker alone, eviction will take
- * care of it. */
- if (pin_inotify_watch(&parent->wdata))
- list_add(&parent->ilist, list);
+ audit_get_parent(parent);
+ fsnotify_destroy_mark(&parent->mark);
+ audit_put_parent(parent);
}
}
}
-/* Update watch data in audit rules based on inotify events. */
-static void audit_handle_ievent(struct inotify_watch *i_watch, u32 wd, u32 mask,
- u32 cookie, const char *dname, struct inode *inode)
+static bool audit_watch_should_send_event(struct fsnotify_group *group, struct inode *inode,
+ struct fsnotify_mark *inode_mark,
+ struct fsnotify_mark *vfsmount_mark,
+ __u32 mask, void *data, int data_type)
+{
+ return true;
+}
+
+/* Update watch data in audit rules based on fsnotify events. */
+static int audit_watch_handle_event(struct fsnotify_group *group,
+ struct fsnotify_mark *inode_mark,
+ struct fsnotify_mark *vfsmount_mark,
+ struct fsnotify_event *event)
{
+ struct inode *inode;
+ __u32 mask = event->mask;
+ const char *dname = event->file_name;
struct audit_parent *parent;
- parent = container_of(i_watch, struct audit_parent, wdata);
+ parent = container_of(inode_mark, struct audit_parent, mark);
- if (mask & (IN_CREATE|IN_MOVED_TO) && inode)
- audit_update_watch(parent, dname, inode->i_sb->s_dev,
- inode->i_ino, 0);
- else if (mask & (IN_DELETE|IN_MOVED_FROM))
+ BUG_ON(group != audit_watch_group);
+
+ switch (event->data_type) {
+ case (FSNOTIFY_EVENT_FILE):
+ inode = event->file->f_path.dentry->d_inode;
+ break;
+ case (FSNOTIFY_EVENT_INODE):
+ inode = event->inode;
+ break;
+ default:
+ BUG();
+ inode = NULL;
+ break;
+ };
+
+ if (mask & (FS_CREATE|FS_MOVED_TO) && inode)
+ audit_update_watch(parent, dname, inode->i_sb->s_dev, inode->i_ino, 0);
+ else if (mask & (FS_DELETE|FS_MOVED_FROM))
audit_update_watch(parent, dname, (dev_t)-1, (unsigned long)-1, 1);
- /* inotify automatically removes the watch and sends IN_IGNORED */
- else if (mask & (IN_DELETE_SELF|IN_UNMOUNT))
- audit_remove_parent_watches(parent);
- /* inotify does not remove the watch, so remove it manually */
- else if(mask & IN_MOVE_SELF) {
+ else if (mask & (FS_DELETE_SELF|FS_UNMOUNT|FS_MOVE_SELF))
audit_remove_parent_watches(parent);
- inotify_remove_watch_locked(audit_ih, i_watch);
- } else if (mask & IN_IGNORED)
- put_inotify_watch(i_watch);
+
+ return 0;
}
-static const struct inotify_operations audit_inotify_ops = {
- .handle_event = audit_handle_ievent,
- .destroy_watch = audit_free_parent,
+static const struct fsnotify_ops audit_watch_fsnotify_ops = {
+ .should_send_event = audit_watch_should_send_event,
+ .handle_event = audit_watch_handle_event,
+ .free_group_priv = NULL,
+ .freeing_mark = NULL,
+ .free_event_priv = NULL,
};
static int __init audit_watch_init(void)
{
- audit_ih = inotify_init(&audit_inotify_ops);
- if (IS_ERR(audit_ih))
- audit_panic("cannot initialize inotify handle");
+ audit_watch_group = fsnotify_alloc_group(&audit_watch_fsnotify_ops);
+ if (IS_ERR(audit_watch_group)) {
+ audit_watch_group = NULL;
+ audit_panic("cannot create audit fsnotify group");
+ }
return 0;
}
-subsys_initcall(audit_watch_init);
+device_initcall(audit_watch_init);
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index ce08041f578..eb7675499fb 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -71,6 +71,7 @@ static inline void audit_free_rule(struct audit_entry *e)
{
int i;
struct audit_krule *erule = &e->rule;
+
/* some rules don't have associated watches */
if (erule->watch)
audit_put_watch(erule->watch);
@@ -746,8 +747,7 @@ static inline int audit_dupe_lsm_field(struct audit_field *df,
* rule with the new rule in the filterlist, then free the old rule.
* The rlist element is undefined; list manipulations are handled apart from
* the initial copy. */
-struct audit_entry *audit_dupe_rule(struct audit_krule *old,
- struct audit_watch *watch)
+struct audit_entry *audit_dupe_rule(struct audit_krule *old)
{
u32 fcount = old->field_count;
struct audit_entry *entry;
@@ -769,8 +769,8 @@ struct audit_entry *audit_dupe_rule(struct audit_krule *old,
new->prio = old->prio;
new->buflen = old->buflen;
new->inode_f = old->inode_f;
- new->watch = NULL;
new->field_count = old->field_count;
+
/*
* note that we are OK with not refcounting here; audit_match_tree()
* never dereferences tree and we can't get false positives there
@@ -811,9 +811,9 @@ struct audit_entry *audit_dupe_rule(struct audit_krule *old,
}
}
- if (watch) {
- audit_get_watch(watch);
- new->watch = watch;
+ if (old->watch) {
+ audit_get_watch(old->watch);
+ new->watch = old->watch;
}
return entry;
@@ -866,7 +866,7 @@ static inline int audit_add_rule(struct audit_entry *entry)
struct audit_watch *watch = entry->rule.watch;
struct audit_tree *tree = entry->rule.tree;
struct list_head *list;
- int h, err;
+ int err;
#ifdef CONFIG_AUDITSYSCALL
int dont_count = 0;
@@ -889,15 +889,11 @@ static inline int audit_add_rule(struct audit_entry *entry)
if (watch) {
/* audit_filter_mutex is dropped and re-taken during this call */
- err = audit_add_watch(&entry->rule);
+ err = audit_add_watch(&entry->rule, &list);
if (err) {
mutex_unlock(&audit_filter_mutex);
goto error;
}
- /* entry->rule.watch may have changed during audit_add_watch() */
- watch = entry->rule.watch;
- h = audit_hash_ino((u32)audit_watch_inode(watch));
- list = &audit_inode_hash[h];
}
if (tree) {
err = audit_add_tree_rule(&entry->rule);
@@ -949,7 +945,6 @@ static inline int audit_del_rule(struct audit_entry *entry)
struct audit_watch *watch = entry->rule.watch;
struct audit_tree *tree = entry->rule.tree;
struct list_head *list;
- LIST_HEAD(inotify_list);
int ret = 0;
#ifdef CONFIG_AUDITSYSCALL
int dont_count = 0;
@@ -969,7 +964,7 @@ static inline int audit_del_rule(struct audit_entry *entry)
}
if (e->rule.watch)
- audit_remove_watch_rule(&e->rule, &inotify_list);
+ audit_remove_watch_rule(&e->rule);
if (e->rule.tree)
audit_remove_tree_rule(&e->rule);
@@ -987,9 +982,6 @@ static inline int audit_del_rule(struct audit_entry *entry)
#endif
mutex_unlock(&audit_filter_mutex);
- if (!list_empty(&inotify_list))
- audit_inotify_unregister(&inotify_list);
-
out:
if (watch)
audit_put_watch(watch); /* match initial get */
@@ -1323,30 +1315,23 @@ static int update_lsm_rule(struct audit_krule *r)
{
struct audit_entry *entry = container_of(r, struct audit_entry, rule);
struct audit_entry *nentry;
- struct audit_watch *watch;
- struct audit_tree *tree;
int err = 0;
if (!security_audit_rule_known(r))
return 0;
- watch = r->watch;
- tree = r->tree;
- nentry = audit_dupe_rule(r, watch);
+ nentry = audit_dupe_rule(r);
if (IS_ERR(nentry)) {
/* save the first error encountered for the
* return value */
err = PTR_ERR(nentry);
audit_panic("error updating LSM filters");
- if (watch)
+ if (r->watch)
list_del(&r->rlist);
list_del_rcu(&entry->list);
list_del(&r->list);
} else {
- if (watch) {
- list_add(&nentry->rule.rlist, audit_watch_rules(watch));
- list_del(&r->rlist);
- } else if (tree)
+ if (r->watch || r->tree)
list_replace_init(&r->rlist, &nentry->rule.rlist);
list_replace_rcu(&entry->list, &nentry->list);
list_replace(&r->list, &nentry->rule.list);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 3828ad5fb8f..b87a63beb66 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -65,7 +65,6 @@
#include <linux/binfmts.h>
#include <linux/highmem.h>
#include <linux/syscalls.h>
-#include <linux/inotify.h>
#include <linux/capability.h>
#include <linux/fs_struct.h>
@@ -549,9 +548,8 @@ static int audit_filter_rules(struct task_struct *tsk,
}
break;
case AUDIT_WATCH:
- if (name && audit_watch_inode(rule->watch) != (unsigned long)-1)
- result = (name->dev == audit_watch_dev(rule->watch) &&
- name->ino == audit_watch_inode(rule->watch));
+ if (name)
+ result = audit_watch_compare(rule->watch, name->ino, name->dev);
break;
case AUDIT_DIR:
if (ctx)
@@ -1726,7 +1724,7 @@ static inline void handle_one(const struct inode *inode)
struct audit_tree_refs *p;
struct audit_chunk *chunk;
int count;
- if (likely(list_empty(&inode->inotify_watches)))
+ if (likely(hlist_empty(&inode->i_fsnotify_marks)))
return;
context = current->audit_context;
p = context->trees;
@@ -1769,7 +1767,7 @@ retry:
seq = read_seqbegin(&rename_lock);
for(;;) {
struct inode *inode = d->d_inode;
- if (inode && unlikely(!list_empty(&inode->inotify_watches))) {
+ if (inode && unlikely(!hlist_empty(&inode->i_fsnotify_marks))) {
struct audit_chunk *chunk;
chunk = audit_tree_lookup(inode);
if (chunk) {
diff --git a/kernel/compat.c b/kernel/compat.c
index 5adab05a317..e167efce842 100644
--- a/kernel/compat.c
+++ b/kernel/compat.c
@@ -279,11 +279,6 @@ asmlinkage long compat_sys_setrlimit(unsigned int resource,
struct compat_rlimit __user *rlim)
{
struct rlimit r;
- int ret;
- mm_segment_t old_fs = get_fs ();
-
- if (resource >= RLIM_NLIMITS)
- return -EINVAL;
if (!access_ok(VERIFY_READ, rlim, sizeof(*rlim)) ||
__get_user(r.rlim_cur, &rlim->rlim_cur) ||
@@ -294,10 +289,7 @@ asmlinkage long compat_sys_setrlimit(unsigned int resource,
r.rlim_cur = RLIM_INFINITY;
if (r.rlim_max == COMPAT_RLIM_INFINITY)
r.rlim_max = RLIM_INFINITY;
- set_fs(KERNEL_DS);
- ret = sys_setrlimit(resource, (struct rlimit __user *) &r);
- set_fs(old_fs);
- return ret;
+ return do_prlimit(current, resource, &r, NULL);
}
#ifdef COMPAT_RLIM_OLD_INFINITY
@@ -329,16 +321,13 @@ asmlinkage long compat_sys_old_getrlimit(unsigned int resource,
#endif
-asmlinkage long compat_sys_getrlimit (unsigned int resource,
+asmlinkage long compat_sys_getrlimit(unsigned int resource,
struct compat_rlimit __user *rlim)
{
struct rlimit r;
int ret;
- mm_segment_t old_fs = get_fs();
- set_fs(KERNEL_DS);
- ret = sys_getrlimit(resource, (struct rlimit __user *) &r);
- set_fs(old_fs);
+ ret = do_prlimit(current, resource, NULL, &r);
if (!ret) {
if (r.rlim_cur > COMPAT_RLIM_INFINITY)
r.rlim_cur = COMPAT_RLIM_INFINITY;
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index f66bdd33a6c..6842eeba587 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -16,13 +16,13 @@
* siglock protection since other code may update expiration cache as
* well.
*/
-void update_rlimit_cpu(unsigned long rlim_new)
+void update_rlimit_cpu(struct task_struct *task, unsigned long rlim_new)
{
cputime_t cputime = secs_to_cputime(rlim_new);
- spin_lock_irq(&current->sighand->siglock);
- set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL);
- spin_unlock_irq(&current->sighand->siglock);
+ spin_lock_irq(&task->sighand->siglock);
+ set_process_cpu_timer(task, CPUCLOCK_PROF, &cputime, NULL);
+ spin_unlock_irq(&task->sighand->siglock);
}
static int check_clock(const clockid_t which_clock)
diff --git a/kernel/power/block_io.c b/kernel/power/block_io.c
index 97024fd40cd..83bbc7c02df 100644
--- a/kernel/power/block_io.c
+++ b/kernel/power/block_io.c
@@ -28,7 +28,7 @@
static int submit(int rw, struct block_device *bdev, sector_t sector,
struct page *page, struct bio **bio_chain)
{
- const int bio_rw = rw | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG);
+ const int bio_rw = rw | REQ_SYNC | REQ_UNPLUG;
struct bio *bio;
bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1);
diff --git a/kernel/sys.c b/kernel/sys.c
index e83ddbbaf89..e9ad4448982 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1236,15 +1236,14 @@ SYSCALL_DEFINE2(setdomainname, char __user *, name, int, len)
SYSCALL_DEFINE2(getrlimit, unsigned int, resource, struct rlimit __user *, rlim)
{
- if (resource >= RLIM_NLIMITS)
- return -EINVAL;
- else {
- struct rlimit value;
- task_lock(current->group_leader);
- value = current->signal->rlim[resource];
- task_unlock(current->group_leader);
- return copy_to_user(rlim, &value, sizeof(*rlim)) ? -EFAULT : 0;
- }
+ struct rlimit value;
+ int ret;
+
+ ret = do_prlimit(current, resource, NULL, &value);
+ if (!ret)
+ ret = copy_to_user(rlim, &value, sizeof(*rlim)) ? -EFAULT : 0;
+
+ return ret;
}
#ifdef __ARCH_WANT_SYS_OLD_GETRLIMIT
@@ -1272,44 +1271,89 @@ SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource,
#endif
-SYSCALL_DEFINE2(setrlimit, unsigned int, resource, struct rlimit __user *, rlim)
+static inline bool rlim64_is_infinity(__u64 rlim64)
{
- struct rlimit new_rlim, *old_rlim;
- int retval;
+#if BITS_PER_LONG < 64
+ return rlim64 >= ULONG_MAX;
+#else
+ return rlim64 == RLIM64_INFINITY;
+#endif
+}
+
+static void rlim_to_rlim64(const struct rlimit *rlim, struct rlimit64 *rlim64)
+{
+ if (rlim->rlim_cur == RLIM_INFINITY)
+ rlim64->rlim_cur = RLIM64_INFINITY;
+ else
+ rlim64->rlim_cur = rlim->rlim_cur;
+ if (rlim->rlim_max == RLIM_INFINITY)
+ rlim64->rlim_max = RLIM64_INFINITY;
+ else
+ rlim64->rlim_max = rlim->rlim_max;
+}
+
+static void rlim64_to_rlim(const struct rlimit64 *rlim64, struct rlimit *rlim)
+{
+ if (rlim64_is_infinity(rlim64->rlim_cur))
+ rlim->rlim_cur = RLIM_INFINITY;
+ else
+ rlim->rlim_cur = (unsigned long)rlim64->rlim_cur;
+ if (rlim64_is_infinity(rlim64->rlim_max))
+ rlim->rlim_max = RLIM_INFINITY;
+ else
+ rlim->rlim_max = (unsigned long)rlim64->rlim_max;
+}
+
+/* make sure you are allowed to change @tsk limits before calling this */
+int do_prlimit(struct task_struct *tsk, unsigned int resource,
+ struct rlimit *new_rlim, struct rlimit *old_rlim)
+{
+ struct rlimit *rlim;
+ int retval = 0;
if (resource >= RLIM_NLIMITS)
return -EINVAL;
- if (copy_from_user(&new_rlim, rlim, sizeof(*rlim)))
- return -EFAULT;
- if (new_rlim.rlim_cur > new_rlim.rlim_max)
- return -EINVAL;
- old_rlim = current->signal->rlim + resource;
- if ((new_rlim.rlim_max > old_rlim->rlim_max) &&
- !capable(CAP_SYS_RESOURCE))
- return -EPERM;
- if (resource == RLIMIT_NOFILE && new_rlim.rlim_max > sysctl_nr_open)
- return -EPERM;
-
- retval = security_task_setrlimit(resource, &new_rlim);
- if (retval)
- return retval;
-
- if (resource == RLIMIT_CPU && new_rlim.rlim_cur == 0) {
- /*
- * The caller is asking for an immediate RLIMIT_CPU
- * expiry. But we use the zero value to mean "it was
- * never set". So let's cheat and make it one second
- * instead
- */
- new_rlim.rlim_cur = 1;
+ if (new_rlim) {
+ if (new_rlim->rlim_cur > new_rlim->rlim_max)
+ return -EINVAL;
+ if (resource == RLIMIT_NOFILE &&
+ new_rlim->rlim_max > sysctl_nr_open)
+ return -EPERM;
}
- task_lock(current->group_leader);
- *old_rlim = new_rlim;
- task_unlock(current->group_leader);
-
- if (resource != RLIMIT_CPU)
+ /* protect tsk->signal and tsk->sighand from disappearing */
+ read_lock(&tasklist_lock);
+ if (!tsk->sighand) {
+ retval = -ESRCH;
goto out;
+ }
+
+ rlim = tsk->signal->rlim + resource;
+ task_lock(tsk->group_leader);
+ if (new_rlim) {
+ if (new_rlim->rlim_max > rlim->rlim_max &&
+ !capable(CAP_SYS_RESOURCE))
+ retval = -EPERM;
+ if (!retval)
+ retval = security_task_setrlimit(tsk->group_leader,
+ resource, new_rlim);
+ if (resource == RLIMIT_CPU && new_rlim->rlim_cur == 0) {
+ /*
+ * The caller is asking for an immediate RLIMIT_CPU
+ * expiry. But we use the zero value to mean "it was
+ * never set". So let's cheat and make it one second
+ * instead
+ */
+ new_rlim->rlim_cur = 1;
+ }
+ }
+ if (!retval) {
+ if (old_rlim)
+ *old_rlim = *rlim;
+ if (new_rlim)
+ *rlim = *new_rlim;
+ }
+ task_unlock(tsk->group_leader);
/*
* RLIMIT_CPU handling. Note that the kernel fails to return an error
@@ -1317,14 +1361,84 @@ SYSCALL_DEFINE2(setrlimit, unsigned int, resource, struct rlimit __user *, rlim)
* very long-standing error, and fixing it now risks breakage of
* applications, so we live with it
*/
- if (new_rlim.rlim_cur == RLIM_INFINITY)
- goto out;
-
- update_rlimit_cpu(new_rlim.rlim_cur);
+ if (!retval && new_rlim && resource == RLIMIT_CPU &&
+ new_rlim->rlim_cur != RLIM_INFINITY)
+ update_rlimit_cpu(tsk, new_rlim->rlim_cur);
out:
+ read_unlock(&tasklist_lock);
+ return retval;
+}
+
+/* rcu lock must be held */
+static int check_prlimit_permission(struct task_struct *task)
+{
+ const struct cred *cred = current_cred(), *tcred;
+
+ tcred = __task_cred(task);
+ if ((cred->uid != tcred->euid ||
+ cred->uid != tcred->suid ||
+ cred->uid != tcred->uid ||
+ cred->gid != tcred->egid ||
+ cred->gid != tcred->sgid ||
+ cred->gid != tcred->gid) &&
+ !capable(CAP_SYS_RESOURCE)) {
+ return -EPERM;
+ }
+
return 0;
}
+SYSCALL_DEFINE4(prlimit64, pid_t, pid, unsigned int, resource,
+ const struct rlimit64 __user *, new_rlim,
+ struct rlimit64 __user *, old_rlim)
+{
+ struct rlimit64 old64, new64;
+ struct rlimit old, new;
+ struct task_struct *tsk;
+ int ret;
+
+ if (new_rlim) {
+ if (copy_from_user(&new64, new_rlim, sizeof(new64)))
+ return -EFAULT;
+ rlim64_to_rlim(&new64, &new);
+ }
+
+ rcu_read_lock();
+ tsk = pid ? find_task_by_vpid(pid) : current;
+ if (!tsk) {
+ rcu_read_unlock();
+ return -ESRCH;
+ }
+ ret = check_prlimit_permission(tsk);
+ if (ret) {
+ rcu_read_unlock();
+ return ret;
+ }
+ get_task_struct(tsk);
+ rcu_read_unlock();
+
+ ret = do_prlimit(tsk, resource, new_rlim ? &new : NULL,
+ old_rlim ? &old : NULL);
+
+ if (!ret && old_rlim) {
+ rlim_to_rlim64(&old, &old64);
+ if (copy_to_user(old_rlim, &old64, sizeof(old64)))
+ ret = -EFAULT;
+ }
+
+ put_task_struct(tsk);
+ return ret;
+}
+
+SYSCALL_DEFINE2(setrlimit, unsigned int, resource, struct rlimit __user *, rlim)
+{
+ struct rlimit new_rlim;
+
+ if (copy_from_user(&new_rlim, rlim, sizeof(*rlim)))
+ return -EFAULT;
+ return do_prlimit(current, resource, &new_rlim, NULL);
+}
+
/*
* It would make sense to put struct rusage in the task_struct,
* except that would make the task_struct be *really big*. After
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 70f2ea758ff..bad369ec540 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -181,3 +181,7 @@ cond_syscall(sys_eventfd2);
/* performance counters: */
cond_syscall(sys_perf_event_open);
+
+/* fanotify! */
+cond_syscall(sys_fanotify_init);
+cond_syscall(sys_fanotify_mark);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 6b005e4912b..ca38e8e3e90 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -44,6 +44,7 @@
#include <linux/times.h>
#include <linux/limits.h>
#include <linux/dcache.h>
+#include <linux/dnotify.h>
#include <linux/syscalls.h>
#include <linux/vmstat.h>
#include <linux/nfs_fs.h>
@@ -131,6 +132,9 @@ static int min_percpu_pagelist_fract = 8;
static int ngroups_max = NGROUPS_MAX;
+#ifdef CONFIG_INOTIFY_USER
+#include <linux/inotify.h>
+#endif
#ifdef CONFIG_SPARC
#include <asm/system.h>
#endif
@@ -207,9 +211,6 @@ static struct ctl_table fs_table[];
static struct ctl_table debug_table[];
static struct ctl_table dev_table[];
extern struct ctl_table random_table[];
-#ifdef CONFIG_INOTIFY_USER
-extern struct ctl_table inotify_table[];
-#endif
#ifdef CONFIG_EPOLL
extern struct ctl_table epoll_table[];
#endif
diff --git a/kernel/timer.c b/kernel/timer.c
index f1b8afe1ad8..97bf05baade 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -326,6 +326,7 @@ EXPORT_SYMBOL_GPL(round_jiffies_up_relative);
/**
* set_timer_slack - set the allowed slack for a timer
+ * @timer: the timer to be modified
* @slack_hz: the amount of time (in jiffies) allowed for rounding
*
* Set the amount of time, in jiffies, that a certain timer has
diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c
index 638711c1750..82499a5bdcb 100644
--- a/kernel/trace/blktrace.c
+++ b/kernel/trace/blktrace.c
@@ -169,9 +169,12 @@ static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector,
static const u32 ddir_act[2] = { BLK_TC_ACT(BLK_TC_READ),
BLK_TC_ACT(BLK_TC_WRITE) };
+#define BLK_TC_HARDBARRIER BLK_TC_BARRIER
+#define BLK_TC_RAHEAD BLK_TC_AHEAD
+
/* The ilog2() calls fall out because they're constant */
-#define MASK_TC_BIT(rw, __name) ((rw & (1 << BIO_RW_ ## __name)) << \
- (ilog2(BLK_TC_ ## __name) + BLK_TC_SHIFT - BIO_RW_ ## __name))
+#define MASK_TC_BIT(rw, __name) ((rw & REQ_ ## __name) << \
+ (ilog2(BLK_TC_ ## __name) + BLK_TC_SHIFT - __REQ_ ## __name))
/*
* The worker for the various blk_add_trace*() types. Fills out a
@@ -194,9 +197,9 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes,
return;
what |= ddir_act[rw & WRITE];
- what |= MASK_TC_BIT(rw, BARRIER);
- what |= MASK_TC_BIT(rw, SYNCIO);
- what |= MASK_TC_BIT(rw, AHEAD);
+ what |= MASK_TC_BIT(rw, HARDBARRIER);
+ what |= MASK_TC_BIT(rw, SYNC);
+ what |= MASK_TC_BIT(rw, RAHEAD);
what |= MASK_TC_BIT(rw, META);
what |= MASK_TC_BIT(rw, DISCARD);
@@ -549,6 +552,41 @@ int blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
}
EXPORT_SYMBOL_GPL(blk_trace_setup);
+#if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64)
+static int compat_blk_trace_setup(struct request_queue *q, char *name,
+ dev_t dev, struct block_device *bdev,
+ char __user *arg)
+{
+ struct blk_user_trace_setup buts;
+ struct compat_blk_user_trace_setup cbuts;
+ int ret;
+
+ if (copy_from_user(&cbuts, arg, sizeof(cbuts)))
+ return -EFAULT;
+
+ buts = (struct blk_user_trace_setup) {
+ .act_mask = cbuts.act_mask,
+ .buf_size = cbuts.buf_size,
+ .buf_nr = cbuts.buf_nr,
+ .start_lba = cbuts.start_lba,
+ .end_lba = cbuts.end_lba,
+ .pid = cbuts.pid,
+ };
+ memcpy(&buts.name, &cbuts.name, 32);
+
+ ret = do_blk_trace_setup(q, name, dev, bdev, &buts);
+ if (ret)
+ return ret;
+
+ if (copy_to_user(arg, &buts.name, 32)) {
+ blk_trace_remove(q);
+ return -EFAULT;
+ }
+
+ return 0;
+}
+#endif
+
int blk_trace_startstop(struct request_queue *q, int start)
{
int ret;
@@ -601,6 +639,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
if (!q)
return -ENXIO;
+ lock_kernel();
mutex_lock(&bdev->bd_mutex);
switch (cmd) {
@@ -608,6 +647,12 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
bdevname(bdev, b);
ret = blk_trace_setup(q, b, bdev->bd_dev, bdev, arg);
break;
+#if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64)
+ case BLKTRACESETUP32:
+ bdevname(bdev, b);
+ ret = compat_blk_trace_setup(q, b, bdev->bd_dev, bdev, arg);
+ break;
+#endif
case BLKTRACESTART:
start = 1;
case BLKTRACESTOP:
@@ -622,6 +667,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
}
mutex_unlock(&bdev->bd_mutex);
+ unlock_kernel();
return ret;
}
@@ -661,10 +707,10 @@ static void blk_add_trace_rq(struct request_queue *q, struct request *rq,
if (likely(!bt))
return;
- if (blk_discard_rq(rq))
- rw |= (1 << BIO_RW_DISCARD);
+ if (rq->cmd_flags & REQ_DISCARD)
+ rw |= REQ_DISCARD;
- if (blk_pc_request(rq)) {
+ if (rq->cmd_type == REQ_TYPE_BLOCK_PC) {
what |= BLK_TC_ACT(BLK_TC_PC);
__blk_add_trace(bt, 0, blk_rq_bytes(rq), rw,
what, rq->errors, rq->cmd_len, rq->cmd);
@@ -925,7 +971,7 @@ void blk_add_driver_data(struct request_queue *q,
if (likely(!bt))
return;
- if (blk_pc_request(rq))
+ if (rq->cmd_type == REQ_TYPE_BLOCK_PC)
__blk_add_trace(bt, 0, blk_rq_bytes(rq), 0,
BLK_TA_DRV_DATA, rq->errors, len, data);
else
@@ -1730,7 +1776,7 @@ void blk_dump_cmd(char *buf, struct request *rq)
int len = rq->cmd_len;
unsigned char *cmd = rq->cmd;
- if (!blk_pc_request(rq)) {
+ if (rq->cmd_type != REQ_TYPE_BLOCK_PC) {
buf[0] = '\0';
return;
}
@@ -1755,20 +1801,20 @@ void blk_fill_rwbs(char *rwbs, u32 rw, int bytes)
if (rw & WRITE)
rwbs[i++] = 'W';
- else if (rw & 1 << BIO_RW_DISCARD)
+ else if (rw & REQ_DISCARD)
rwbs[i++] = 'D';
else if (bytes)
rwbs[i++] = 'R';
else
rwbs[i++] = 'N';
- if (rw & 1 << BIO_RW_AHEAD)
+ if (rw & REQ_RAHEAD)
rwbs[i++] = 'A';
- if (rw & 1 << BIO_RW_BARRIER)
+ if (rw & REQ_HARDBARRIER)
rwbs[i++] = 'B';
- if (rw & 1 << BIO_RW_SYNCIO)
+ if (rw & REQ_SYNC)
rwbs[i++] = 'S';
- if (rw & 1 << BIO_RW_META)
+ if (rw & REQ_META)
rwbs[i++] = 'M';
rwbs[i] = '\0';
@@ -1779,8 +1825,8 @@ void blk_fill_rwbs_rq(char *rwbs, struct request *rq)
int rw = rq->cmd_flags & 0x03;
int bytes;
- if (blk_discard_rq(rq))
- rw |= (1 << BIO_RW_DISCARD);
+ if (rq->cmd_flags & REQ_DISCARD)
+ rw |= REQ_DISCARD;
bytes = blk_rq_bytes(rq);