Merge master.kernel.org:/pub/scm/linux/kernel/git/dwmw2/audit-2.6

10 files changed, 356 insertions, 171 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index a67bf7d315d7..d1e0eb46d201 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -370,7 +370,10 @@ W:	http://atmelwlandriver.sourceforge.net/
 S:	Maintained
 
 AUDIT SUBSYSTEM
-L:	linux-audit@redhat.com (subscribers-only)
+P:	David Woodhouse
+M:	dwmw2@infradead.org
+L:	linux-audit@redhat.com
+W:	http://people.redhat.com/sgrubb/audit/
 S:	Maintained
 
 AX.25 NETWORK LAYER
diff --git a/arch/ppc64/kernel/asm-offsets.c b/arch/ppc64/kernel/asm-offsets.c
index 17e35d0fed09..1ff4fa05a973 100644
--- a/arch/ppc64/kernel/asm-offsets.c
+++ b/arch/ppc64/kernel/asm-offsets.c
@@ -68,6 +68,7 @@ int main(void)
 	DEFINE(THREAD_USED_VR, offsetof(struct thread_struct, used_vr));
 #endif /* CONFIG_ALTIVEC */
 	DEFINE(MM, offsetof(struct task_struct, mm));
+	DEFINE(AUDITCONTEXT, offsetof(struct task_struct, audit_context));
 
 	DEFINE(DCACHEL1LINESIZE, offsetof(struct ppc64_caches, dline_size));
 	DEFINE(DCACHEL1LOGLINESIZE, offsetof(struct ppc64_caches, log_dline_size));
diff --git a/arch/ppc64/kernel/entry.S b/arch/ppc64/kernel/entry.S
index d133a49cdf89..e8c0bbf4d000 100644
--- a/arch/ppc64/kernel/entry.S
+++ b/arch/ppc64/kernel/entry.S
@@ -276,12 +276,22 @@ _GLOBAL(ppc64_rt_sigsuspend)
 _GLOBAL(ppc32_rt_sigsuspend)
 	bl	.save_nvgprs
 	bl	.sys32_rt_sigsuspend
-	/* If sigsuspend() returns zero, we are going into a signal handler */
 70:	cmpdi	0,r3,0
-	beq	.ret_from_except
-	/* If it returned -EINTR, we need to return via syscall_exit to set
+	/* If it returned an error, we need to return via syscall_exit to set
 	   the SO bit in cr0 and potentially stop for ptrace. */
-	b	syscall_exit
+	bne	syscall_exit
+	/* If sigsuspend() returns zero, we are going into a signal handler. We
+	   may need to call audit_syscall_exit() to mark the exit from sigsuspend() */
+#ifdef CONFIG_AUDIT
+	ld	r3,PACACURRENT(r13)
+	ld	r4,AUDITCONTEXT(r3)
+	cmpdi	0,r4,0
+	beq	.ret_from_except	/* No audit_context: Leave immediately. */
+	li	r4, 2			/* AUDITSC_FAILURE */
+	li	r5,-4			/* It's always -EINTR */
+	bl	.audit_syscall_exit
+#endif
+	b	.ret_from_except
 
 _GLOBAL(ppc_fork)
 	bl	.save_nvgprs
diff --git a/fs/namei.c b/fs/namei.c
index 21d85f1ac839..043d587216b5 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1048,7 +1048,7 @@ int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata
 out:
 	if (unlikely(current->audit_context
 		     && nd && nd->dentry && nd->dentry->d_inode))
-		audit_inode(name, nd->dentry->d_inode);
+		audit_inode(name, nd->dentry->d_inode, flags);
 	return retval;
 }
 
diff --git a/include/linux/audit.h b/include/linux/audit.h
index 68aba0c02e49..b2a2509bd7ea 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -51,7 +51,8 @@
 #define AUDIT_WATCH_LIST	1009	/* List all file/dir watches */
 #define AUDIT_SIGNAL_INFO	1010	/* Get info about sender of signal to auditd */
 
-#define AUDIT_FIRST_USER_MSG	1100	/* Userspace messages uninteresting to kernel */
+#define AUDIT_FIRST_USER_MSG	1100	/* Userspace messages mostly uninteresting to kernel */
+#define AUDIT_USER_AVC		1107	/* We filter this differently */
 #define AUDIT_LAST_USER_MSG	1199
  
 #define AUDIT_DAEMON_START      1200    /* Daemon startup record */
@@ -75,10 +76,15 @@
 #define AUDIT_KERNEL		2000	/* Asynchronous audit record. NOT A REQUEST. */
 
 /* Rule flags */
-#define AUDIT_PER_TASK 0x01	/* Apply rule at task creation (not syscall) */
-#define AUDIT_AT_ENTRY 0x02	/* Apply rule at syscall entry */
-#define AUDIT_AT_EXIT  0x04	/* Apply rule at syscall exit */
-#define AUDIT_PREPEND  0x10	/* Prepend to front of list */
+#define AUDIT_FILTER_USER	0x00	/* Apply rule to user-generated messages */
+#define AUDIT_FILTER_TASK	0x01	/* Apply rule at task creation (not syscall) */
+#define AUDIT_FILTER_ENTRY	0x02	/* Apply rule at syscall entry */
+#define AUDIT_FILTER_WATCH	0x03	/* Apply rule to file system watches */
+#define AUDIT_FILTER_EXIT	0x04	/* Apply rule at syscall exit */
+
+#define AUDIT_NR_FILTERS	5
+
+#define AUDIT_FILTER_PREPEND	0x10	/* Prepend to front of list */
 
 /* Rule actions */
 #define AUDIT_NEVER    0	/* Do not build context if rule matches */
@@ -199,6 +205,7 @@ struct audit_sig_info {
 struct audit_buffer;
 struct audit_context;
 struct inode;
+struct netlink_skb_parms;
 
 #define AUDITSC_INVALID 0
 #define AUDITSC_SUCCESS 1
@@ -215,7 +222,7 @@ extern void audit_syscall_entry(struct task_struct *task, int arch,
 extern void audit_syscall_exit(struct task_struct *task, int failed, long return_code);
 extern void audit_getname(const char *name);
 extern void audit_putname(const char *name);
-extern void audit_inode(const char *name, const struct inode *inode);
+extern void audit_inode(const char *name, const struct inode *inode, unsigned flags);
 
 				/* Private API (for audit.c only) */
 extern int  audit_receive_filter(int type, int pid, int uid, int seq,
@@ -230,6 +237,7 @@ extern int audit_socketcall(int nargs, unsigned long *args);
 extern int audit_sockaddr(int len, void *addr);
 extern int audit_avc_path(struct dentry *dentry, struct vfsmount *mnt);
 extern void audit_signal_info(int sig, struct task_struct *t);
+extern int audit_filter_user(struct netlink_skb_parms *cb, int type);
 #else
 #define audit_alloc(t) ({ 0; })
 #define audit_free(t) do { ; } while (0)
@@ -237,7 +245,7 @@ extern void audit_signal_info(int sig, struct task_struct *t);
 #define audit_syscall_exit(t,f,r) do { ; } while (0)
 #define audit_getname(n) do { ; } while (0)
 #define audit_putname(n) do { ; } while (0)
-#define audit_inode(n,i) do { ; } while (0)
+#define audit_inode(n,i,f) do { ; } while (0)
 #define audit_receive_filter(t,p,u,s,d,l) ({ -EOPNOTSUPP; })
 #define auditsc_get_stamp(c,t,s) do { BUG(); } while (0)
 #define audit_get_loginuid(c) ({ -1; })
@@ -246,16 +254,17 @@ extern void audit_signal_info(int sig, struct task_struct *t);
 #define audit_sockaddr(len, addr) ({ 0; })
 #define audit_avc_path(dentry, mnt) ({ 0; })
 #define audit_signal_info(s,t) do { ; } while (0)
+#define audit_filter_user(cb,t) ({ 1; })
 #endif
 
 #ifdef CONFIG_AUDIT
 /* These are defined in audit.c */
 				/* Public API */
-extern void		    audit_log(struct audit_context *ctx, int type,
-				      const char *fmt, ...)
-			    __attribute__((format(printf,3,4)));
+extern void		    audit_log(struct audit_context *ctx, int gfp_mask,
+				      int type, const char *fmt, ...)
+				      __attribute__((format(printf,4,5)));
 
-extern struct audit_buffer *audit_log_start(struct audit_context *ctx,int type);
+extern struct audit_buffer *audit_log_start(struct audit_context *ctx, int gfp_mask, int type);
 extern void		    audit_log_format(struct audit_buffer *ab,
 					     const char *fmt, ...)
 			    __attribute__((format(printf,2,3)));
@@ -274,9 +283,10 @@ extern void		    audit_send_reply(int pid, int seq, int type,
 					     int done, int multi,
 					     void *payload, int size);
 extern void		    audit_log_lost(const char *message);
+extern struct semaphore audit_netlink_sem;
 #else
-#define audit_log(c,t,f,...) do { ; } while (0)
-#define audit_log_start(c,t) ({ NULL; })
+#define audit_log(c,g,t,f,...) do { ; } while (0)
+#define audit_log_start(c,g,t) ({ NULL; })
 #define audit_log_vformat(b,f,a) do { ; } while (0)
 #define audit_log_format(b,f,...) do { ; } while (0)
 #define audit_log_end(b) do { ; } while (0)
diff --git a/kernel/audit.c b/kernel/audit.c
index 7f0699790d46..83096b67510a 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -79,6 +79,8 @@ static int	audit_rate_limit;
 
 /* Number of outstanding audit_buffers allowed. */
 static int	audit_backlog_limit = 64;
+static int	audit_backlog_wait_time = 60 * HZ;
+static int	audit_backlog_wait_overflow = 0;
 
 /* The identity of the user shutting down the audit system. */
 uid_t		audit_sig_uid = -1;
@@ -106,18 +108,12 @@ static LIST_HEAD(audit_freelist);
 static struct sk_buff_head audit_skb_queue;
 static struct task_struct *kauditd_task;
 static DECLARE_WAIT_QUEUE_HEAD(kauditd_wait);
-
-/* There are three lists of rules -- one to search at task creation
- * time, one to search at syscall entry time, and another to search at
- * syscall exit time. */
-static LIST_HEAD(audit_tsklist);
-static LIST_HEAD(audit_entlist);
-static LIST_HEAD(audit_extlist);
+static DECLARE_WAIT_QUEUE_HEAD(audit_backlog_wait);
 
 /* The netlink socket is only to be read by 1 CPU, which lets us assume
  * that list additions and deletions never happen simultaneously in
  * auditsc.c */
-static DECLARE_MUTEX(audit_netlink_sem);
+DECLARE_MUTEX(audit_netlink_sem);
 
 /* AUDIT_BUFSIZ is the size of the temporary buffer used for formatting
  * audit records.  Since printk uses a 1024 byte buffer, this buffer
@@ -137,6 +133,7 @@ struct audit_buffer {
 	struct list_head     list;
 	struct sk_buff       *skb;	/* formatted skb ready to send */
 	struct audit_context *ctx;	/* NULL or associated context */
+	int		     gfp_mask;
 };
 
 static void audit_set_pid(struct audit_buffer *ab, pid_t pid)
@@ -145,11 +142,6 @@ static void audit_set_pid(struct audit_buffer *ab, pid_t pid)
 	nlh->nlmsg_pid = pid;
 }
 
-struct audit_entry {
-	struct list_head  list;
-	struct audit_rule rule;
-};
-
 static void audit_panic(const char *message)
 {
 	switch (audit_failure)
@@ -233,7 +225,7 @@ static int audit_set_rate_limit(int limit, uid_t loginuid)
 {
 	int old		 = audit_rate_limit;
 	audit_rate_limit = limit;
-	audit_log(NULL, AUDIT_CONFIG_CHANGE, 
+	audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE, 
 			"audit_rate_limit=%d old=%d by auid=%u",
 			audit_rate_limit, old, loginuid);
 	return old;
@@ -243,7 +235,7 @@ static int audit_set_backlog_limit(int limit, uid_t loginuid)
 {
 	int old		 = audit_backlog_limit;
 	audit_backlog_limit = limit;
-	audit_log(NULL, AUDIT_CONFIG_CHANGE,
+	audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
 			"audit_backlog_limit=%d old=%d by auid=%u",
 			audit_backlog_limit, old, loginuid);
 	return old;
@@ -255,7 +247,7 @@ static int audit_set_enabled(int state, uid_t loginuid)
 	if (state != 0 && state != 1)
 		return -EINVAL;
 	audit_enabled = state;
-	audit_log(NULL, AUDIT_CONFIG_CHANGE,
+	audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
 			"audit_enabled=%d old=%d by auid=%u",
 			audit_enabled, old, loginuid);
 	return old;
@@ -269,7 +261,7 @@ static int audit_set_failure(int state, uid_t loginuid)
 	    && state != AUDIT_FAIL_PANIC)
 		return -EINVAL;
 	audit_failure = state;
-	audit_log(NULL, AUDIT_CONFIG_CHANGE,
+	audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
 			"audit_failure=%d old=%d by auid=%u",
 			audit_failure, old, loginuid);
 	return old;
@@ -281,6 +273,7 @@ int kauditd_thread(void *dummy)
 
 	while (1) {
 		skb = skb_dequeue(&audit_skb_queue);
+		wake_up(&audit_backlog_wait);
 		if (skb) {
 			if (audit_pid) {
 				int err = netlink_unicast(audit_sock, skb, audit_pid, 0);
@@ -290,7 +283,7 @@ int kauditd_thread(void *dummy)
 					audit_pid = 0;
 				}
 			} else {
-				printk(KERN_ERR "%s\n", skb->data + NLMSG_SPACE(0));
+				printk(KERN_NOTICE "%s\n", skb->data + NLMSG_SPACE(0));
 				kfree_skb(skb);
 			}
 		} else {
@@ -423,7 +416,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		if (status_get->mask & AUDIT_STATUS_PID) {
 			int old   = audit_pid;
 			audit_pid = status_get->pid;
-			audit_log(NULL, AUDIT_CONFIG_CHANGE,
+			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
 				"audit_pid=%d old=%d by auid=%u",
 				  audit_pid, old, loginuid);
 		}
@@ -435,15 +428,21 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		break;
 	case AUDIT_USER:
 	case AUDIT_FIRST_USER_MSG...AUDIT_LAST_USER_MSG:
-		ab = audit_log_start(NULL, msg_type);
-		if (!ab)
-			break;	/* audit_panic has been called */
-		audit_log_format(ab,
-				 "user pid=%d uid=%u auid=%u"
-				 " msg='%.1024s'",
-				 pid, uid, loginuid, (char *)data);
-		audit_set_pid(ab, pid);
-		audit_log_end(ab);
+		if (!audit_enabled && msg_type != AUDIT_USER_AVC)
+			return 0;
+
+		err = audit_filter_user(&NETLINK_CB(skb), msg_type);
+		if (err == 1) {
+			err = 0;
+			ab = audit_log_start(NULL, GFP_KERNEL, msg_type);
+			if (ab) {
+				audit_log_format(ab,
+						 "user pid=%d uid=%u auid=%u msg='%.1024s'",
+						 pid, uid, loginuid, (char *)data);
+				audit_set_pid(ab, pid);
+				audit_log_end(ab);
+			}
+		}
 		break;
 	case AUDIT_ADD:
 	case AUDIT_DEL:
@@ -523,7 +522,7 @@ static int __init audit_init(void)
 	skb_queue_head_init(&audit_skb_queue);
 	audit_initialized = 1;
 	audit_enabled = audit_default;
-	audit_log(NULL, AUDIT_KERNEL, "initialized");
+	audit_log(NULL, GFP_KERNEL, AUDIT_KERNEL, "initialized");
 	return 0;
 }
 __initcall(audit_init);
@@ -561,7 +560,7 @@ static void audit_buffer_free(struct audit_buffer *ab)
 }
 
 static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx,
-						int gfp_mask, int type)
+						unsigned int __nocast gfp_mask, int type)
 {
 	unsigned long flags;
 	struct audit_buffer *ab = NULL;
@@ -587,6 +586,7 @@ static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx,
 		goto err;
 
 	ab->ctx = ctx;
+	ab->gfp_mask = gfp_mask;
 	nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0));
 	nlh->nlmsg_type = type;
 	nlh->nlmsg_flags = 0;
@@ -606,26 +606,27 @@ err:
  * (timestamp,serial) tuple is unique for each syscall and is live from
  * syscall entry to syscall exit.
  *
- * Atomic values are only guaranteed to be 24-bit, so we count down.
- *
  * NOTE: Another possibility is to store the formatted records off the
  * audit context (for those records that have a context), and emit them
  * all at syscall exit.  However, this could delay the reporting of
  * significant errors until syscall exit (or never, if the system
  * halts). */
+
 unsigned int audit_serial(void)
 {
-	static atomic_t serial = ATOMIC_INIT(0xffffff);
-	unsigned int a, b;
+	static spinlock_t serial_lock = SPIN_LOCK_UNLOCKED;
+	static unsigned int serial = 0;
+
+	unsigned long flags;
+	unsigned int ret;
 
+	spin_lock_irqsave(&serial_lock, flags);
 	do {
-		a = atomic_read(&serial);
-		if (atomic_dec_and_test(&serial))
-			atomic_set(&serial, 0xffffff);
-		b = atomic_read(&serial);
-	} while (b != a - 1);
+		ret = ++serial;
+	} while (unlikely(!ret));
+	spin_unlock_irqrestore(&serial_lock, flags);
 
-	return 0xffffff - b;
+	return ret;
 }
 
 static inline void audit_get_stamp(struct audit_context *ctx, 
@@ -645,17 +646,43 @@ static inline void audit_get_stamp(struct audit_context *ctx,
  * syscall, then the syscall is marked as auditable and an audit record
  * will be written at syscall exit.  If there is no associated task, tsk
  * should be NULL. */
-struct audit_buffer *audit_log_start(struct audit_context *ctx, int type)
+
+struct audit_buffer *audit_log_start(struct audit_context *ctx, int gfp_mask,
+				     int type)
 {
 	struct audit_buffer	*ab	= NULL;
 	struct timespec		t;
 	unsigned int		serial;
+	int reserve;
+	unsigned long timeout_start = jiffies;
 
 	if (!audit_initialized)
 		return NULL;
 
-	if (audit_backlog_limit
-	    && skb_queue_len(&audit_skb_queue) > audit_backlog_limit) {
+	if (gfp_mask & __GFP_WAIT)
+		reserve = 0;
+	else
+		reserve = 5; /* Allow atomic callers to go up to five 
+				entries over the normal backlog limit */
+
+	while (audit_backlog_limit
+	       && skb_queue_len(&audit_skb_queue) > audit_backlog_limit + reserve) {
+		if (gfp_mask & __GFP_WAIT && audit_backlog_wait_time
+		    && time_before(jiffies, timeout_start + audit_backlog_wait_time)) {
+
+			/* Wait for auditd to drain the queue a little */
+			DECLARE_WAITQUEUE(wait, current);
+			set_current_state(TASK_INTERRUPTIBLE);
+			add_wait_queue(&audit_backlog_wait, &wait);
+
+			if (audit_backlog_limit &&
+			    skb_queue_len(&audit_skb_queue) > audit_backlog_limit)
+				schedule_timeout(timeout_start + audit_backlog_wait_time - jiffies);
+
+			__set_current_state(TASK_RUNNING);
+			remove_wait_queue(&audit_backlog_wait, &wait);
+			continue;
+		}
 		if (audit_rate_check())
 			printk(KERN_WARNING
 			       "audit: audit_backlog=%d > "
@@ -663,10 +690,12 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, int type)
 			       skb_queue_len(&audit_skb_queue),
 			       audit_backlog_limit);
 		audit_log_lost("backlog limit exceeded");
+		audit_backlog_wait_time = audit_backlog_wait_overflow;
+		wake_up(&audit_backlog_wait);
 		return NULL;
 	}
 
-	ab = audit_buffer_alloc(ctx, GFP_ATOMIC, type);
+	ab = audit_buffer_alloc(ctx, gfp_mask, type);
 	if (!ab) {
 		audit_log_lost("out of memory in audit_log_start");
 		return NULL;
@@ -690,7 +719,7 @@ static inline int audit_expand(struct audit_buffer *ab, int extra)
 {
 	struct sk_buff *skb = ab->skb;
 	int ret = pskb_expand_head(skb, skb_headroom(skb), extra,
-				   GFP_ATOMIC);
+				   ab->gfp_mask);
 	if (ret < 0) {
 		audit_log_lost("out of memory in audit_expand");
 		return 0;
@@ -809,7 +838,7 @@ void audit_log_d_path(struct audit_buffer *ab, const char *prefix,
 		audit_log_format(ab, " %s", prefix);
 
 	/* We will allow 11 spaces for ' (deleted)' to be appended */
-	path = kmalloc(PATH_MAX+11, GFP_KERNEL);
+	path = kmalloc(PATH_MAX+11, ab->gfp_mask);
 	if (!path) {
 		audit_log_format(ab, "<no memory>");
 		return;
@@ -841,7 +870,7 @@ void audit_log_end(struct audit_buffer *ab)
 			ab->skb = NULL;
 			wake_up_interruptible(&kauditd_wait);
 		} else {
-			printk("%s\n", ab->skb->data + NLMSG_SPACE(0));
+			printk(KERN_NOTICE "%s\n", ab->skb->data + NLMSG_SPACE(0));
 		}
 	}
 	audit_buffer_free(ab);
@@ -850,12 +879,13 @@ void audit_log_end(struct audit_buffer *ab)
 /* Log an audit record.  This is a convenience function that calls
  * audit_log_start, audit_log_vformat, and audit_log_end.  It may be
  * called in any context. */
-void audit_log(struct audit_context *ctx, int type, const char *fmt, ...)
+void audit_log(struct audit_context *ctx, int gfp_mask, int type, 
+	       const char *fmt, ...)
 {
 	struct audit_buffer *ab;
 	va_list args;
 
-	ab = audit_log_start(ctx, type);
+	ab = audit_log_start(ctx, gfp_mask, type);
 	if (ab) {
 		va_start(args, fmt);
 		audit_log_vformat(ab, fmt, args);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index e75f84e1a1a0..88696f639aab 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -39,6 +39,9 @@
 #include <linux/audit.h>
 #include <linux/personality.h>
 #include <linux/time.h>
+#include <linux/kthread.h>
+#include <linux/netlink.h>
+#include <linux/compiler.h>
 #include <asm/unistd.h>
 
 /* 0 = no checking
@@ -95,6 +98,7 @@ struct audit_names {
 	uid_t		uid;
 	gid_t		gid;
 	dev_t		rdev;
+	unsigned	flags;
 };
 
 struct audit_aux_data {
@@ -167,9 +171,16 @@ struct audit_context {
 /* There are three lists of rules -- one to search at task creation
  * time, one to search at syscall entry time, and another to search at
  * syscall exit time. */
-static LIST_HEAD(audit_tsklist);
-static LIST_HEAD(audit_entlist);
-static LIST_HEAD(audit_extlist);
+static struct list_head audit_filter_list[AUDIT_NR_FILTERS] = {
+	LIST_HEAD_INIT(audit_filter_list[0]),
+	LIST_HEAD_INIT(audit_filter_list[1]),
+	LIST_HEAD_INIT(audit_filter_list[2]),
+	LIST_HEAD_INIT(audit_filter_list[3]),
+	LIST_HEAD_INIT(audit_filter_list[4]),
+#if AUDIT_NR_FILTERS != 5
+#error Fix audit_filter_list initialiser
+#endif
+};
 
 struct audit_entry {
 	struct list_head  list;
@@ -179,9 +190,36 @@ struct audit_entry {
 
 extern int audit_pid;
 
+/* Copy rule from user-space to kernel-space.  Called from 
+ * audit_add_rule during AUDIT_ADD. */
+static inline int audit_copy_rule(struct audit_rule *d, struct audit_rule *s)
+{
+	int i;
+
+	if (s->action != AUDIT_NEVER
+	    && s->action != AUDIT_POSSIBLE
+	    && s->action != AUDIT_ALWAYS)
+		return -1;
+	if (s->field_count < 0 || s->field_count > AUDIT_MAX_FIELDS)
+		return -1;
+	if ((s->flags & ~AUDIT_FILTER_PREPEND) >= AUDIT_NR_FILTERS)
+		return -1;
+
+	d->flags	= s->flags;
+	d->action	= s->action;
+	d->field_count	= s->field_count;
+	for (i = 0; i < d->field_count; i++) {
+		d->fields[i] = s->fields[i];
+		d->values[i] = s->values[i];
+	}
+	for (i = 0; i < AUDIT_BITMASK_SIZE; i++) d->mask[i] = s->mask[i];
+	return 0;
+}
+
 /* Check to see if two rules are identical.  It is called from
+ * audit_add_rule during AUDIT_ADD and 
  * audit_del_rule during AUDIT_DEL. */
-static int audit_compare_rule(struct audit_rule *a, struct audit_rule *b)
+static inline int audit_compare_rule(struct audit_rule *a, struct audit_rule *b)
 {
 	int i;
 
@@ -210,19 +248,37 @@ static int audit_compare_rule(struct audit_rule *a, struct audit_rule *b)
 /* Note that audit_add_rule and audit_del_rule are called via
  * audit_receive() in audit.c, and are protected by
  * audit_netlink_sem. */
-static inline int audit_add_rule(struct audit_entry *entry,
-				 struct list_head *list)
+static inline int audit_add_rule(struct audit_rule *rule,
+				  struct list_head *list)
 {
-	if (entry->rule.flags & AUDIT_PREPEND) {
-		entry->rule.flags &= ~AUDIT_PREPEND;
+	struct audit_entry  *entry;
+
+	/* Do not use the _rcu iterator here, since this is the only
+	 * addition routine. */
+	list_for_each_entry(entry, list, list) {
+		if (!audit_compare_rule(rule, &entry->rule)) {
+			return -EEXIST;
+		}
+	}
+
+	if (!(entry = kmalloc(sizeof(*entry), GFP_KERNEL)))
+		return -ENOMEM;
+	if (audit_copy_rule(&entry->rule, rule)) {
+		kfree(entry);
+		return -EINVAL;
+	}
+
+	if (entry->rule.flags & AUDIT_FILTER_PREPEND) {
+		entry->rule.flags &= ~AUDIT_FILTER_PREPEND;
 		list_add_rcu(&entry->list, list);
 	} else {
 		list_add_tail_rcu(&entry->list, list);
 	}
+
 	return 0;
 }
 
-static void audit_free_rule(struct rcu_head *head)
+static inline void audit_free_rule(struct rcu_head *head)
 {
 	struct audit_entry *e = container_of(head, struct audit_entry, rcu);
 	kfree(e);
@@ -245,82 +301,82 @@ static inline int audit_del_rule(struct audit_rule *rule,
 			return 0;
 		}
 	}
-	return -EFAULT;		/* No matching rule */
+	return -ENOENT;		/* No matching rule */
 }
 
-/* Copy rule from user-space to kernel-space.  Called during
- * AUDIT_ADD. */
-static int audit_copy_rule(struct audit_rule *d, struct audit_rule *s)
+static int audit_list_rules(void *_dest)
 {
+	int pid, seq;
+	int *dest = _dest;
+	struct audit_entry *entry;
 	int i;
 
-	if (s->action != AUDIT_NEVER
-	    && s->action != AUDIT_POSSIBLE
-	    && s->action != AUDIT_ALWAYS)
-		return -1;
-	if (s->field_count < 0 || s->field_count > AUDIT_MAX_FIELDS)
-		return -1;
+	pid = dest[0];
+	seq = dest[1];
+	kfree(dest);
 
-	d->flags	= s->flags;
-	d->action	= s->action;
-	d->field_count	= s->field_count;
-	for (i = 0; i < d->field_count; i++) {
-		d->fields[i] = s->fields[i];
-		d->values[i] = s->values[i];
+	down(&audit_netlink_sem);
+
+	/* The *_rcu iterators not needed here because we are
+	   always called with audit_netlink_sem held. */
+	for (i=0; i<AUDIT_NR_FILTERS; i++) {
+		list_for_each_entry(entry, &audit_filter_list[i], list)
+			audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
+					 &entry->rule, sizeof(entry->rule));
 	}
-	for (i = 0; i < AUDIT_BITMASK_SIZE; i++) d->mask[i] = s->mask[i];
+	audit_send_reply(pid, seq, AUDIT_LIST, 1, 1, NULL, 0);
+	
+	up(&audit_netlink_sem);
 	return 0;
 }
 
 int audit_receive_filter(int type, int pid, int uid, int seq, void *data,
 							uid_t loginuid)
 {
-	u32		   flags;
-	struct audit_entry *entry;
+	struct task_struct *tsk;
+	int *dest;
 	int		   err = 0;
+	unsigned listnr;
 
 	switch (type) {
 	case AUDIT_LIST:
-		/* The *_rcu iterators not needed here because we are
-		   always called with audit_netlink_sem held. */
-		list_for_each_entry(entry, &audit_tsklist, list)
-			audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
-					 &entry->rule, sizeof(entry->rule));
-		list_for_each_entry(entry, &audit_entlist, list)
-			audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
-					 &entry->rule, sizeof(entry->rule));
-		list_for_each_entry(entry, &audit_extlist, list)
-			audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
-					 &entry->rule, sizeof(entry->rule));
-		audit_send_reply(pid, seq, AUDIT_LIST, 1, 1, NULL, 0);
+		/* We can't just spew out the rules here because we might fill
+		 * the available socket buffer space and deadlock waiting for
+		 * auditctl to read from it... which isn't ever going to
+		 * happen if we're actually running in the context of auditctl
+		 * trying to _send_ the stuff */
+		 
+		dest = kmalloc(2 * sizeof(int), GFP_KERNEL);
+		if (!dest)
+			return -ENOMEM;
+		dest[0] = pid;
+		dest[1] = seq;
+
+		tsk = kthread_run(audit_list_rules, dest, "audit_list_rules");
+		if (IS_ERR(tsk)) {
+			kfree(dest);
+			err = PTR_ERR(tsk);
+		}
 		break;
 	case AUDIT_ADD:
-		if (!(entry = kmalloc(sizeof(*entry), GFP_KERNEL)))
-			return -ENOMEM;
-		if (audit_copy_rule(&entry->rule, data)) {
-			kfree(entry);
+		listnr =((struct audit_rule *)data)->flags & ~AUDIT_FILTER_PREPEND;
+		if (listnr >= AUDIT_NR_FILTERS)
 			return -EINVAL;
-		}
-		flags = entry->rule.flags;
-		if (!err && (flags & AUDIT_PER_TASK))
-			err = audit_add_rule(entry, &audit_tsklist);
-		if (!err && (flags & AUDIT_AT_ENTRY))
-			err = audit_add_rule(entry, &audit_entlist);
-		if (!err && (flags & AUDIT_AT_EXIT))
-			err = audit_add_rule(entry, &audit_extlist);
-		audit_log(NULL, AUDIT_CONFIG_CHANGE, 
-				"auid=%u added an audit rule\n", loginuid);
+
+		err = audit_add_rule(data, &audit_filter_list[listnr]);
+		if (!err)
+			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
+				  "auid=%u added an audit rule\n", loginuid);
 		break;
 	case AUDIT_DEL:
-		flags =((struct audit_rule *)data)->flags;
-		if (!err && (flags & AUDIT_PER_TASK))
-			err = audit_del_rule(data, &audit_tsklist);
-		if (!err && (flags & AUDIT_AT_ENTRY))
-			err = audit_del_rule(data, &audit_entlist);
-		if (!err && (flags & AUDIT_AT_EXIT))
-			err = audit_del_rule(data, &audit_extlist);
-		audit_log(NULL, AUDIT_CONFIG_CHANGE,
-				"auid=%u removed an audit rule\n", loginuid);
+		listnr =((struct audit_rule *)data)->flags & ~AUDIT_FILTER_PREPEND;
+		if (listnr >= AUDIT_NR_FILTERS)
+			return -EINVAL;
+
+		err = audit_del_rule(data, &audit_filter_list[listnr]);
+		if (!err)
+			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
+				  "auid=%u removed an audit rule\n", loginuid);
 		break;
 	default:
 		return -EINVAL;
@@ -384,8 +440,12 @@ static int audit_filter_rules(struct task_struct *tsk,
 				result = (ctx->return_code == value);
 			break;
 		case AUDIT_SUCCESS:
-			if (ctx && ctx->return_valid)
-				result = (ctx->return_valid == AUDITSC_SUCCESS);
+			if (ctx && ctx->return_valid) {
+				if (value)
+					result = (ctx->return_valid == AUDITSC_SUCCESS);
+				else
+					result = (ctx->return_valid == AUDITSC_FAILURE);
+			}
 			break;
 		case AUDIT_DEVMAJOR:
 			if (ctx) {
@@ -454,7 +514,7 @@ static enum audit_state audit_filter_task(struct task_struct *tsk)
 	enum audit_state   state;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(e, &audit_tsklist, list) {
+	list_for_each_entry_rcu(e, &audit_filter_list[AUDIT_FILTER_TASK], list) {
 		if (audit_filter_rules(tsk, &e->rule, NULL, &state)) {
 			rcu_read_unlock();
 			return state;
@@ -474,20 +534,84 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk,
 					     struct list_head *list)
 {
 	struct audit_entry *e;
+	enum audit_state state;
+
+	if (audit_pid && tsk->tgid == audit_pid)
+		return AUDIT_DISABLED;
+
+	rcu_read_lock();
+	if (!list_empty(list)) {
+		    int word = AUDIT_WORD(ctx->major);
+		    int bit  = AUDIT_BIT(ctx->major);
+
+		    list_for_each_entry_rcu(e, list, list) {
+			    if ((e->rule.mask[word] & bit) == bit
+				&& audit_filter_rules(tsk, &e->rule, ctx, &state)) {
+				    rcu_read_unlock();
+				    return state;
+			    }
+		    }
+	}
+	rcu_read_unlock();
+	return AUDIT_BUILD_CONTEXT;
+}
+
+static int audit_filter_user_rules(struct netlink_skb_parms *cb,
+			      struct audit_rule *rule,
+			      enum audit_state *state)
+{
+	int i;
+
+	for (i = 0; i < rule->field_count; i++) {
+		u32 field  = rule->fields[i] & ~AUDIT_NEGATE;
+		u32 value  = rule->values[i];
+		int result = 0;
+
+		switch (field) {
+		case AUDIT_PID:
+			result = (cb->creds.pid == value);
+			break;
+		case AUDIT_UID:
+			result = (cb->creds.uid == value);
+			break;
+		case AUDIT_GID:
+			result = (cb->creds.gid == value);
+			break;
+		case AUDIT_LOGINUID:
+			result = (cb->loginuid == value);
+			break;
+		}
+
+		if (rule->fields[i] & AUDIT_NEGATE)
+			result = !result;
+		if (!result)
+			return 0;
+	}
+	switch (rule->action) {
+	case AUDIT_NEVER:    *state = AUDIT_DISABLED;	    break;
+	case AUDIT_POSSIBLE: *state = AUDIT_BUILD_CONTEXT;  break;
+	case AUDIT_ALWAYS:   *state = AUDIT_RECORD_CONTEXT; break;
+	}
+	return 1;
+}
+
+int audit_filter_user(struct netlink_skb_parms *cb, int type)
+{
+	struct audit_entry *e;
 	enum audit_state   state;
-	int		   word = AUDIT_WORD(ctx->major);
-	int		   bit  = AUDIT_BIT(ctx->major);
+	int ret = 1;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(e, list, list) {
-		if ((e->rule.mask[word] & bit) == bit
- 		    && audit_filter_rules(tsk, &e->rule, ctx, &state)) {
-			rcu_read_unlock();
-			return state;
+	list_for_each_entry_rcu(e, &audit_filter_list[AUDIT_FILTER_USER], list) {
+		if (audit_filter_user_rules(cb, &e->rule, &state)) {
+			if (state == AUDIT_DISABLED)
+				ret = 0;
+			break;
 		}
 	}
 	rcu_read_unlock();
-	return AUDIT_BUILD_CONTEXT;
+
+	return ret; /* Audit by default */
 }
 
 /* This should be called with task_lock() held. */
@@ -504,7 +628,7 @@ static inline struct audit_context *audit_get_context(struct task_struct *tsk,
 
 	if (context->in_syscall && !context->auditable) {
 		enum audit_state state;
-		state = audit_filter_syscall(tsk, context, &audit_extlist);
+		state = audit_filter_syscall(tsk, context, &audit_filter_list[AUDIT_FILTER_EXIT]);
 		if (state == AUDIT_RECORD_CONTEXT)
 			context->auditable = 1;
 	}
@@ -679,13 +803,13 @@ static void audit_log_task_info(struct audit_buffer *ab)
 	up_read(&mm->mmap_sem);
 }
 
-static void audit_log_exit(struct audit_context *context)
+static void audit_log_exit(struct audit_context *context, unsigned int gfp_mask)
 {
 	int i;
 	struct audit_buffer *ab;
 	struct audit_aux_data *aux;
 
-	ab = audit_log_start(context, AUDIT_SYSCALL);
+	ab = audit_log_start(context, gfp_mask, AUDIT_SYSCALL);
 	if (!ab)
 		return;		/* audit_panic has been called */
 	audit_log_format(ab, "arch=%x syscall=%d",
@@ -717,7 +841,7 @@ static void audit_log_exit(struct audit_context *context)
 
 	for (aux = context->aux; aux; aux = aux->next) {
 
-		ab = audit_log_start(context, aux->type);
+		ab = audit_log_start(context, GFP_KERNEL, aux->type);
 		if (!ab)
 			continue; /* audit_panic has been called */
 
@@ -754,14 +878,14 @@ static void audit_log_exit(struct audit_context *context)
 	}
 
 	if (context->pwd && context->pwdmnt) {
-		ab = audit_log_start(context, AUDIT_CWD);
+		ab = audit_log_start(context, GFP_KERNEL, AUDIT_CWD);
 		if (ab) {
 			audit_log_d_path(ab, "cwd=", context->pwd, context->pwdmnt);
 			audit_log_end(ab);
 		}
 	}
 	for (i = 0; i < context->name_count; i++) {
-		ab = audit_log_start(context, AUDIT_PATH);
+		ab = audit_log_start(context, GFP_KERNEL, AUDIT_PATH);
 		if (!ab)
 			continue; /* audit_panic has been called */
 
@@ -770,6 +894,8 @@ static void audit_log_exit(struct audit_context *context)
 			audit_log_format(ab, " name=");
 			audit_log_untrustedstring(ab, context->names[i].name);
 		}
+		audit_log_format(ab, " flags=%x\n", context->names[i].flags);
+			 
 		if (context->names[i].ino != (unsigned long)-1)
 			audit_log_format(ab, " inode=%lu dev=%02x:%02x mode=%#o"
 					     " ouid=%u ogid=%u rdev=%02x:%02x",
@@ -799,9 +925,11 @@ void audit_free(struct task_struct *tsk)
 		return;
 
 	/* Check for system calls that do not go through the exit
-	 * function (e.g., exit_group), then free context block. */
-	if (context->in_syscall && context->auditable && context->pid != audit_pid)
-		audit_log_exit(context);
+	 * function (e.g., exit_group), then free context block. 
+	 * We use GFP_ATOMIC here because we might be doing this 
+	 * in the context of the idle thread */
+	if (context->in_syscall && context->auditable)
+		audit_log_exit(context, GFP_ATOMIC);
 
 	audit_free_context(context);
 }
@@ -876,11 +1004,11 @@ void audit_syscall_entry(struct task_struct *tsk, int arch, int major,
 
 	state = context->state;
 	if (state == AUDIT_SETUP_CONTEXT || state == AUDIT_BUILD_CONTEXT)
-		state = audit_filter_syscall(tsk, context, &audit_entlist);
+		state = audit_filter_syscall(tsk, context, &audit_filter_list[AUDIT_FILTER_ENTRY]);
 	if (likely(state == AUDIT_DISABLED))
 		return;
 
-	context->serial     = audit_serial();
+	context->serial     = 0;
 	context->ctime      = CURRENT_TIME;
 	context->in_syscall = 1;
 	context->auditable  = !!(state == AUDIT_RECORD_CONTEXT);
@@ -903,10 +1031,10 @@ void audit_syscall_exit(struct task_struct *tsk, int valid, long return_code)
 	/* Not having a context here is ok, since the parent may have
 	 * called __put_task_struct. */
 	if (likely(!context))
-		return;
+		goto out;
 
-	if (context->in_syscall && context->auditable && context->pid != audit_pid)
-		audit_log_exit(context);
+	if (context->in_syscall && context->auditable)
+		audit_log_exit(context, GFP_KERNEL);
 
 	context->in_syscall = 0;
 	context->auditable  = 0;
@@ -919,9 +1047,9 @@ void audit_syscall_exit(struct task_struct *tsk, int valid, long return_code)
 	} else {
 		audit_free_names(context);
 		audit_free_aux(context);
-		audit_zero_context(context, context->state);
 		tsk->audit_context = context;
 	}
+ out:
 	put_task_struct(tsk);
 }
 
@@ -996,7 +1124,7 @@ void audit_putname(const char *name)
 
 /* Store the inode and device from a lookup.  Called from
  * fs/namei.c:path_lookup(). */
-void audit_inode(const char *name, const struct inode *inode)
+void audit_inode(const char *name, const struct inode *inode, unsigned flags)
 {
 	int idx;
 	struct audit_context *context = current->audit_context;
@@ -1022,17 +1150,20 @@ void audit_inode(const char *name, const struct inode *inode)
 		++context->ino_count;
 #endif
 	}
-	context->names[idx].ino  = inode->i_ino;
-	context->names[idx].dev	 = inode->i_sb->s_dev;
-	context->names[idx].mode = inode->i_mode;
-	context->names[idx].uid  = inode->i_uid;
-	context->names[idx].gid  = inode->i_gid;
-	context->names[idx].rdev = inode->i_rdev;
+	context->names[idx].flags = flags;
+	context->names[idx].ino   = inode->i_ino;
+	context->names[idx].dev	  = inode->i_sb->s_dev;
+	context->names[idx].mode  = inode->i_mode;
+	context->names[idx].uid   = inode->i_uid;
+	context->names[idx].gid   = inode->i_gid;
+	context->names[idx].rdev  = inode->i_rdev;
 }
 
 void auditsc_get_stamp(struct audit_context *ctx,
 		       struct timespec *t, unsigned int *serial)
 {
+	if (!ctx->serial)
+		ctx->serial = audit_serial();
 	t->tv_sec  = ctx->ctime.tv_sec;
 	t->tv_nsec = ctx->ctime.tv_nsec;
 	*serial    = ctx->serial;
@@ -1044,7 +1175,7 @@ int audit_set_loginuid(struct task_struct *task, uid_t loginuid)
 	if (task->audit_context) {
 		struct audit_buffer *ab;
 
-		ab = audit_log_start(NULL, AUDIT_LOGIN);
+		ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_LOGIN);
 		if (ab) {
 			audit_log_format(ab, "login pid=%d uid=%u "
 				"old auid=%u new auid=%u",
@@ -1153,7 +1284,7 @@ void audit_signal_info(int sig, struct task_struct *t)
 	extern pid_t audit_sig_pid;
 	extern uid_t audit_sig_uid;
 
-	if (unlikely(audit_pid && t->pid == audit_pid)) {
+	if (unlikely(audit_pid && t->tgid == audit_pid)) {
 		if (sig == SIGTERM || sig == SIGHUP) {
 			struct audit_context *ctx = current->audit_context;
 			audit_sig_pid = current->pid;
diff --git a/security/selinux/avc.c b/security/selinux/avc.c
index cf6020f85403..12e4fb72bf0f 100644
--- a/security/selinux/avc.c
+++ b/security/selinux/avc.c
@@ -242,7 +242,7 @@ void __init avc_init(void)
 	avc_node_cachep = kmem_cache_create("avc_node", sizeof(struct avc_node),
 					     0, SLAB_PANIC, NULL, NULL);
 
-	audit_log(current->audit_context, AUDIT_KERNEL, "AVC INITIALIZED\n");
+	audit_log(current->audit_context, GFP_KERNEL, AUDIT_KERNEL, "AVC INITIALIZED\n");
 }
 
 int avc_get_hash_stats(char *page)
@@ -550,7 +550,7 @@ void avc_audit(u32 ssid, u32 tsid,
 			return;
 	}
 
-	ab = audit_log_start(current->audit_context, AUDIT_AVC);
+	ab = audit_log_start(current->audit_context, GFP_ATOMIC, AUDIT_AVC);
 	if (!ab)
 		return;		/* audit_panic has been called */
 	audit_log_format(ab, "avc:  %s ", denied ? "denied" : "granted");
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index f40c8221ec1b..6e4937fe062b 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3389,7 +3389,7 @@ static int selinux_nlmsg_perm(struct sock *sk, struct sk_buff *skb)
 	err = selinux_nlmsg_lookup(isec->sclass, nlh->nlmsg_type, &perm);
 	if (err) {
 		if (err == -EINVAL) {
-			audit_log(current->audit_context, AUDIT_SELINUX_ERR,
+			audit_log(current->audit_context, GFP_KERNEL, AUDIT_SELINUX_ERR,
 				  "SELinux:  unrecognized netlink message"
 				  " type=%hu for sclass=%hu\n",
 				  nlh->nlmsg_type, isec->sclass);
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 92b89dc99bcd..aecdded55e74 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -381,7 +381,7 @@ static int security_validtrans_handle_fail(struct context *ocontext,
 		goto out;
 	if (context_struct_to_string(tcontext, &t, &tlen) < 0)
 		goto out;
-	audit_log(current->audit_context, AUDIT_SELINUX_ERR,
+	audit_log(current->audit_context, GFP_ATOMIC, AUDIT_SELINUX_ERR,
 	          "security_validate_transition:  denied for"
 	          " oldcontext=%s newcontext=%s taskcontext=%s tclass=%s",
 	          o, n, t, policydb.p_class_val_to_name[tclass-1]);
@@ -787,7 +787,7 @@ static int compute_sid_handle_invalid_context(
 		goto out;
 	if (context_struct_to_string(newcontext, &n, &nlen) < 0)
 		goto out;
-	audit_log(current->audit_context, AUDIT_SELINUX_ERR,
+	audit_log(current->audit_context, GFP_ATOMIC, AUDIT_SELINUX_ERR,
 		  "security_compute_sid:  invalid context %s"
 		  " for scontext=%s"
 		  " tcontext=%s"