Skip to content

L
linux
Commit 28a3a7eb authored by Eric Paris's avatar Eric Paris
Browse files
Options

audit: reimplement audit_trees using fsnotify rather than inotify 

Simply switch audit_trees from using inotify to using fsnotify for it's
inode pinning and disappearing act information.
Signed-off-by: default avatarEric Paris <eparis@redhat.com>
parent 40554c3d
Show whitespace changes
Inline Side-by-side
Showing with 136 additions and 109 deletions
include/linux/fsnotify_backend.h
View file @ 28a3a7eb
...@@ -63,7 +63,8 @@ ...@@ -63,7 +63,8 @@
/* listeners that hard code group numbers near the top */ /* listeners that hard code group numbers near the top */
#define DNOTIFY_GROUP_NUM UINT_MAX #define DNOTIFY_GROUP_NUM UINT_MAX
#define AUDIT_WATCH_GROUP_NUM (DNOTIFY_GROUP_NUM-1) #define AUDIT_WATCH_GROUP_NUM (DNOTIFY_GROUP_NUM-1)
#define INOTIFY_GROUP_NUM (AUDIT_WATCH_GROUP_NUM-1) #define AUDIT_TREE_GROUP_NUM (AUDIT_WATCH_GROUP_NUM-1)
#define INOTIFY_GROUP_NUM (AUDIT_TREE_GROUP_NUM-1)
struct fsnotify_group; struct fsnotify_group;
struct fsnotify_event; struct fsnotify_event;
......
init/Kconfig
View file @ 28a3a7eb
...@@ -326,7 +326,7 @@ config AUDITSYSCALL ...@@ -326,7 +326,7 @@ config AUDITSYSCALL
config AUDIT_TREE config AUDIT_TREE
def_bool y def_bool y
depends on AUDITSYSCALL depends on AUDITSYSCALL
select INOTIFY select FSNOTIFY
menu "RCU Subsystem" menu "RCU Subsystem"
......
kernel/audit_tree.c
View file @ 28a3a7eb
#include "audit.h" #include "audit.h"
#include <linux/inotify.h> #include <linux/fsnotify_backend.h>
#include <linux/namei.h> #include <linux/namei.h>
#include <linux/mount.h> #include <linux/mount.h>
#include <linux/kthread.h> #include <linux/kthread.h>
...@@ -22,7 +22,7 @@ struct audit_tree { ...@@ -22,7 +22,7 @@ struct audit_tree {
struct audit_chunk { struct audit_chunk {
struct list_head hash; struct list_head hash;
struct inotify_watch watch; struct fsnotify_mark_entry mark;
struct list_head trees; /* with root here */ struct list_head trees; /* with root here */
int dead; int dead;
int count; int count;
...@@ -59,7 +59,7 @@ static LIST_HEAD(prune_list); ...@@ -59,7 +59,7 @@ static LIST_HEAD(prune_list);
* tree is refcounted; one reference for "some rules on rules_list refer to * tree is refcounted; one reference for "some rules on rules_list refer to
* it", one for each chunk with pointer to it. * 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). * of watch contributes 1 to .refs).
* *
* node.index allows to get from node.list to containing chunk. * node.index allows to get from node.list to containing chunk.
...@@ -68,7 +68,7 @@ static LIST_HEAD(prune_list); ...@@ -68,7 +68,7 @@ static LIST_HEAD(prune_list);
* that makes a difference. Some. * 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) static struct audit_tree *alloc_tree(const char *s)
{ {
...@@ -111,29 +111,6 @@ const char *audit_tree_path(struct audit_tree *tree) ...@@ -111,29 +111,6 @@ const char *audit_tree_path(struct audit_tree *tree)
return tree->pathname; 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) static void free_chunk(struct audit_chunk *chunk)
{ {
int i; int i;
...@@ -157,6 +134,35 @@ static void __put_chunk(struct rcu_head *rcu) ...@@ -157,6 +134,35 @@ static void __put_chunk(struct rcu_head *rcu)
audit_put_chunk(chunk); audit_put_chunk(chunk);
} }
static void audit_tree_destroy_watch(struct fsnotify_mark_entry *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}; enum {HASH_SIZE = 128};
static struct list_head chunk_hash_heads[HASH_SIZE]; static struct list_head chunk_hash_heads[HASH_SIZE];
static __cacheline_aligned_in_smp DEFINE_SPINLOCK(hash_lock); static __cacheline_aligned_in_smp DEFINE_SPINLOCK(hash_lock);
...@@ -167,10 +173,15 @@ static inline struct list_head *chunk_hash(const struct inode *inode) ...@@ -167,10 +173,15 @@ static inline struct list_head *chunk_hash(const struct inode *inode)
return chunk_hash_heads + n % HASH_SIZE; 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) static void insert_hash(struct audit_chunk *chunk)
{ {
struct list_head *list = chunk_hash(chunk->watch.inode); struct fsnotify_mark_entry *entry = &chunk->mark;
struct list_head *list;
if (!entry->inode)
return;
list = chunk_hash(entry->inode);
list_add_rcu(&chunk->hash, list); list_add_rcu(&chunk->hash, list);
} }
...@@ -181,7 +192,8 @@ struct audit_chunk *audit_tree_lookup(const struct inode *inode) ...@@ -181,7 +192,8 @@ struct audit_chunk *audit_tree_lookup(const struct inode *inode)
struct audit_chunk *p; struct audit_chunk *p;
list_for_each_entry_rcu(p, list, hash) { 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.inode == inode) {
atomic_long_inc(&p->refs); atomic_long_inc(&p->refs);
return p; return p;
} }
...@@ -210,38 +222,19 @@ static struct audit_chunk *find_chunk(struct node *p) ...@@ -210,38 +222,19 @@ static struct audit_chunk *find_chunk(struct node *p)
static void untag_chunk(struct node *p) static void untag_chunk(struct node *p)
{ {
struct audit_chunk *chunk = find_chunk(p); struct audit_chunk *chunk = find_chunk(p);
struct fsnotify_mark_entry *entry = &chunk->mark;
struct audit_chunk *new; struct audit_chunk *new;
struct audit_tree *owner; struct audit_tree *owner;
int size = chunk->count - 1; int size = chunk->count - 1;
int i, j; int i, j;
if (!pin_inotify_watch(&chunk->watch)) { fsnotify_get_mark(entry);
/*
* 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;
}
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
/* spin_lock(&entry->lock);
* pin_inotify_watch() succeeded, so the watch won't go away if (chunk->dead || !entry->inode) {
* from under us. spin_unlock(&entry->lock);
*/
mutex_lock(&chunk->watch.inode->inotify_mutex);
if (chunk->dead) {
mutex_unlock(&chunk->watch.inode->inotify_mutex);
goto out; goto out;
} }
...@@ -256,16 +249,17 @@ static void untag_chunk(struct node *p) ...@@ -256,16 +249,17 @@ static void untag_chunk(struct node *p)
list_del_init(&p->list); list_del_init(&p->list);
list_del_rcu(&chunk->hash); list_del_rcu(&chunk->hash);
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
inotify_evict_watch(&chunk->watch); spin_unlock(&entry->lock);
mutex_unlock(&chunk->watch.inode->inotify_mutex); fsnotify_destroy_mark_by_entry(entry);
put_inotify_watch(&chunk->watch); fsnotify_put_mark(entry);
goto out; goto out;
} }
new = alloc_chunk(size); new = alloc_chunk(size);
if (!new) if (!new)
goto Fallback; 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.inode, 1)) {
free_chunk(new); free_chunk(new);
goto Fallback; goto Fallback;
} }
...@@ -298,9 +292,9 @@ static void untag_chunk(struct node *p) ...@@ -298,9 +292,9 @@ static void untag_chunk(struct node *p)
list_for_each_entry(owner, &new->trees, same_root) list_for_each_entry(owner, &new->trees, same_root)
owner->root = new; owner->root = new;
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
inotify_evict_watch(&chunk->watch); spin_unlock(&entry->lock);
mutex_unlock(&chunk->watch.inode->inotify_mutex); fsnotify_destroy_mark_by_entry(entry);
put_inotify_watch(&chunk->watch); fsnotify_put_mark(entry);
goto out; goto out;
Fallback: Fallback:
...@@ -314,31 +308,33 @@ static void untag_chunk(struct node *p) ...@@ -314,31 +308,33 @@ static void untag_chunk(struct node *p)
p->owner = NULL; p->owner = NULL;
put_tree(owner); put_tree(owner);
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
mutex_unlock(&chunk->watch.inode->inotify_mutex); spin_unlock(&entry->lock);
out: out:
unpin_inotify_watch(&chunk->watch); fsnotify_put_mark(entry);
spin_lock(&hash_lock); spin_lock(&hash_lock);
} }
static int create_chunk(struct inode *inode, struct audit_tree *tree) static int create_chunk(struct inode *inode, struct audit_tree *tree)
{ {
struct fsnotify_mark_entry *entry;
struct audit_chunk *chunk = alloc_chunk(1); struct audit_chunk *chunk = alloc_chunk(1);
if (!chunk) if (!chunk)
return -ENOMEM; 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, 0)) {
free_chunk(chunk); free_chunk(chunk);
return -ENOSPC; return -ENOSPC;
} }
mutex_lock(&inode->inotify_mutex); spin_lock(&entry->lock);
spin_lock(&hash_lock); spin_lock(&hash_lock);
if (tree->goner) { if (tree->goner) {
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
chunk->dead = 1; chunk->dead = 1;
inotify_evict_watch(&chunk->watch); spin_unlock(&entry->lock);
mutex_unlock(&inode->inotify_mutex); fsnotify_destroy_mark_by_entry(entry);
put_inotify_watch(&chunk->watch); fsnotify_put_mark(entry);
return 0; return 0;
} }
chunk->owners[0].index = (1U << 31); chunk->owners[0].index = (1U << 31);
...@@ -351,30 +347,33 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) ...@@ -351,30 +347,33 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree)
} }
insert_hash(chunk); insert_hash(chunk);
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
mutex_unlock(&inode->inotify_mutex); spin_unlock(&entry->lock);
return 0; return 0;
} }
/* the first tagged inode becomes root of tree */ /* the first tagged inode becomes root of tree */
static int tag_chunk(struct inode *inode, struct audit_tree *tree) static int tag_chunk(struct inode *inode, struct audit_tree *tree)
{ {
struct inotify_watch *watch; struct fsnotify_mark_entry *old_entry, *chunk_entry;
struct audit_tree *owner; struct audit_tree *owner;
struct audit_chunk *chunk, *old; struct audit_chunk *chunk, *old;
struct node *p; struct node *p;
int n; int n;
if (inotify_find_watch(rtree_ih, inode, &watch) < 0) spin_lock(&inode->i_lock);
old_entry = fsnotify_find_mark_entry(audit_tree_group, inode);
spin_unlock(&inode->i_lock);
if (!old_entry)
return create_chunk(inode, tree); 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? */ /* are we already there? */
spin_lock(&hash_lock); spin_lock(&hash_lock);
for (n = 0; n < old->count; n++) { for (n = 0; n < old->count; n++) {
if (old->owners[n].owner == tree) { if (old->owners[n].owner == tree) {
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
put_inotify_watch(&old->watch); fsnotify_put_mark(old_entry);
return 0; return 0;
} }
} }
...@@ -382,25 +381,44 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) ...@@ -382,25 +381,44 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree)
chunk = alloc_chunk(old->count + 1); chunk = alloc_chunk(old->count + 1);
if (!chunk) { if (!chunk) {
put_inotify_watch(&old->watch); fsnotify_put_mark(old_entry);
return -ENOMEM; return -ENOMEM;
} }
mutex_lock(&inode->inotify_mutex); chunk_entry = &chunk->mark;
if (inotify_clone_watch(&old->watch, &chunk->watch) < 0) {
mutex_unlock(&inode->inotify_mutex); spin_lock(&old_entry->lock);
put_inotify_watch(&old->watch); if (!old_entry->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->inode, 1)) {
spin_unlock(&old_entry->lock);
free_chunk(chunk); free_chunk(chunk);
fsnotify_put_mark(old_entry);
return -ENOSPC; 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); spin_lock(&hash_lock);
/* we now hold old_entry->lock, chunk_entry->lock, and hash_lock */
if (tree->goner) { if (tree->goner) {
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
chunk->dead = 1; chunk->dead = 1;
inotify_evict_watch(&chunk->watch); spin_unlock(&chunk_entry->lock);
mutex_unlock(&inode->inotify_mutex); spin_unlock(&old_entry->lock);
put_inotify_watch(&old->watch);
put_inotify_watch(&chunk->watch); fsnotify_destroy_mark_by_entry(chunk_entry);
fsnotify_put_mark(chunk_entry);
fsnotify_put_mark(old_entry);
return 0; return 0;
} }
list_replace_init(&old->trees, &chunk->trees); list_replace_init(&old->trees, &chunk->trees);
...@@ -426,10 +444,11 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) ...@@ -426,10 +444,11 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree)
list_add(&tree->same_root, &chunk->trees); list_add(&tree->same_root, &chunk->trees);
} }
spin_unlock(&hash_lock); spin_unlock(&hash_lock);
inotify_evict_watch(&old->watch); spin_unlock(&chunk_entry->lock);
mutex_unlock(&inode->inotify_mutex); spin_unlock(&old_entry->lock);
put_inotify_watch(&old->watch); /* pair to inotify_find_watch */ fsnotify_destroy_mark_by_entry(old_entry);
put_inotify_watch(&old->watch); /* and kill it */ fsnotify_put_mark(old_entry); /* pair to fsnotify_find mark_entry */
fsnotify_put_mark(old_entry); /* and kill it */
return 0; return 0;
} }
...@@ -584,7 +603,9 @@ void audit_trim_trees(void) ...@@ -584,7 +603,9 @@ void audit_trim_trees(void)
spin_lock(&hash_lock); spin_lock(&hash_lock);
list_for_each_entry(node, &tree->chunks, list) { 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.inode;
node->index |= 1U<<31; node->index |= 1U<<31;
if (iterate_mounts(compare_root, inode, root_mnt)) if (iterate_mounts(compare_root, inode, root_mnt))
node->index &= ~(1U<<31); node->index &= ~(1U<<31);
...@@ -846,7 +867,6 @@ void audit_kill_trees(struct list_head *list) ...@@ -846,7 +867,6 @@ void audit_kill_trees(struct list_head *list)
* Here comes the stuff asynchronous to auditctl operations * Here comes the stuff asynchronous to auditctl operations
*/ */
/* inode->inotify_mutex is locked */
static void evict_chunk(struct audit_chunk *chunk) static void evict_chunk(struct audit_chunk *chunk)
{ {
struct audit_tree *owner; struct audit_tree *owner;
...@@ -885,35 +905,41 @@ static void evict_chunk(struct audit_chunk *chunk) ...@@ -885,35 +905,41 @@ static void evict_chunk(struct audit_chunk *chunk)
mutex_unlock(&audit_filter_mutex); mutex_unlock(&audit_filter_mutex);
} }
static void handle_event(struct inotify_watch *watch, u32 wd, u32 mask, static int audit_tree_handle_event(struct fsnotify_group *group, struct fsnotify_event *event)
u32 cookie, const char *dname, struct inode *inode)
{ {
struct audit_chunk *chunk = container_of(watch, struct audit_chunk, watch); BUG();
return -EOPNOTSUPP;
}
static void audit_tree_freeing_mark(struct fsnotify_mark_entry *entry, struct fsnotify_group *group)
{
struct audit_chunk *chunk = container_of(entry, struct audit_chunk, mark);
if (mask & IN_IGNORED) {
evict_chunk(chunk); evict_chunk(chunk);
put_inotify_watch(watch); 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, __u32 mask)
{ {
struct audit_chunk *chunk = container_of(watch, struct audit_chunk, watch); return 0;
call_rcu(&chunk->head, __put_chunk);
} }
static const struct inotify_operations rtree_inotify_ops = { static const struct fsnotify_ops audit_tree_ops = {
.handle_event = handle_event, .handle_event = audit_tree_handle_event,
.destroy_watch = destroy_watch, .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) static int __init audit_tree_init(void)
{ {
int i; int i;
rtree_ih = inotify_init(&rtree_inotify_ops); audit_tree_group = fsnotify_obtain_group(AUDIT_TREE_GROUP_NUM,
if (IS_ERR(rtree_ih)) 0, &audit_tree_ops);
audit_panic("cannot initialize inotify handle for rectree watches"); if (IS_ERR(audit_tree_group))
audit_panic("cannot initialize fsnotify group for rectree watches");
for (i = 0; i < HASH_SIZE; i++) for (i = 0; i < HASH_SIZE; i++)
INIT_LIST_HEAD(&chunk_hash_heads[i]); INIT_LIST_HEAD(&chunk_hash_heads[i]);
......
kernel/auditsc.c
View file @ 28a3a7eb
...@@ -1725,7 +1725,7 @@ static inline void handle_one(const struct inode *inode) ...@@ -1725,7 +1725,7 @@ static inline void handle_one(const struct inode *inode)
struct audit_tree_refs *p; struct audit_tree_refs *p;
struct audit_chunk *chunk; struct audit_chunk *chunk;
int count; int count;
if (likely(list_empty(&inode->inotify_watches))) if (likely(hlist_empty(&inode->i_fsnotify_mark_entries)))
return; return;
context = current->audit_context; context = current->audit_context;
p = context->trees; p = context->trees;
...@@ -1768,7 +1768,7 @@ static void handle_path(const struct dentry *dentry) ...@@ -1768,7 +1768,7 @@ static void handle_path(const struct dentry *dentry)
seq = read_seqbegin(&rename_lock); seq = read_seqbegin(&rename_lock);
for(;;) { for(;;) {
struct inode *inode = d->d_inode; struct inode *inode = d->d_inode;
if (inode && unlikely(!list_empty(&inode->inotify_watches))) { if (inode && unlikely(!hlist_empty(&inode->i_fsnotify_mark_entries))) {
struct audit_chunk *chunk; struct audit_chunk *chunk;
chunk = audit_tree_lookup(inode); chunk = audit_tree_lookup(inode);
if (chunk) { if (chunk) {
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7