1 files changed, 208 insertions, 70 deletions

diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c
index ea719cdd6a36..3a3d05cfe09a 100644
--- a/fs/lockd/svclock.c
+++ b/fs/lockd/svclock.c
@@ -30,7 +30,6 @@
 #include <linux/sunrpc/svc_xprt.h>
 #include <linux/lockd/nlm.h>
 #include <linux/lockd/lockd.h>
-#include <linux/kthread.h>
 
 #define NLMDBG_FACILITY		NLMDBG_SVCLOCK
 
@@ -130,12 +129,14 @@ static void nlmsvc_insert_block(struct nlm_block *block, unsigned long when)
 static inline void
 nlmsvc_remove_block(struct nlm_block *block)
 {
+	spin_lock(&nlm_blocked_lock);
 	if (!list_empty(&block->b_list)) {
-		spin_lock(&nlm_blocked_lock);
 		list_del_init(&block->b_list);
 		spin_unlock(&nlm_blocked_lock);
 		nlmsvc_release_block(block);
+		return;
 	}
+	spin_unlock(&nlm_blocked_lock);
 }
 
 /*
@@ -148,21 +149,25 @@ nlmsvc_lookup_block(struct nlm_file *file, struct nlm_lock *lock)
 	struct file_lock	*fl;
 
 	dprintk("lockd: nlmsvc_lookup_block f=%p pd=%d %Ld-%Ld ty=%d\n",
-				file, lock->fl.fl_pid,
+				file, lock->fl.c.flc_pid,
 				(long long)lock->fl.fl_start,
-				(long long)lock->fl.fl_end, lock->fl.fl_type);
+				(long long)lock->fl.fl_end,
+				lock->fl.c.flc_type);
+	spin_lock(&nlm_blocked_lock);
 	list_for_each_entry(block, &nlm_blocked, b_list) {
 		fl = &block->b_call->a_args.lock.fl;
 		dprintk("lockd: check f=%p pd=%d %Ld-%Ld ty=%d cookie=%s\n",
-				block->b_file, fl->fl_pid,
+				block->b_file, fl->c.flc_pid,
 				(long long)fl->fl_start,
-				(long long)fl->fl_end, fl->fl_type,
+				(long long)fl->fl_end, fl->c.flc_type,
 				nlmdbg_cookie2a(&block->b_call->a_args.cookie));
 		if (block->b_file == file && nlm_compare_locks(fl, &lock->fl)) {
 			kref_get(&block->b_count);
+			spin_unlock(&nlm_blocked_lock);
 			return block;
 		}
 	}
+	spin_unlock(&nlm_blocked_lock);
 
 	return NULL;
 }
@@ -184,16 +189,19 @@ nlmsvc_find_block(struct nlm_cookie *cookie)
 {
 	struct nlm_block *block;
 
+	spin_lock(&nlm_blocked_lock);
 	list_for_each_entry(block, &nlm_blocked, b_list) {
 		if (nlm_cookie_match(&block->b_call->a_args.cookie,cookie))
 			goto found;
 	}
+	spin_unlock(&nlm_blocked_lock);
 
 	return NULL;
 
 found:
 	dprintk("nlmsvc_find_block(%s): block=%p\n", nlmdbg_cookie2a(cookie), block);
 	kref_get(&block->b_count);
+	spin_unlock(&nlm_blocked_lock);
 	return block;
 }
 
@@ -236,7 +244,7 @@ nlmsvc_create_block(struct svc_rqst *rqstp, struct nlm_host *host,
 		goto failed_free;
 
 	/* Set notifier function for VFS, and init args */
-	call->a_args.lock.fl.fl_flags |= FL_SLEEP;
+	call->a_args.lock.fl.c.flc_flags |= FL_SLEEP;
 	call->a_args.lock.fl.fl_lmops = &nlmsvc_lock_operations;
 	nlmclnt_next_cookie(&call->a_args.cookie);
 
@@ -316,6 +324,7 @@ void nlmsvc_traverse_blocks(struct nlm_host *host,
 
 restart:
 	mutex_lock(&file->f_mutex);
+	spin_lock(&nlm_blocked_lock);
 	list_for_each_entry_safe(block, next, &file->f_blocks, b_flist) {
 		if (!match(block->b_host, host))
 			continue;
@@ -324,14 +333,85 @@ restart:
 		if (list_empty(&block->b_list))
 			continue;
 		kref_get(&block->b_count);
+		spin_unlock(&nlm_blocked_lock);
 		mutex_unlock(&file->f_mutex);
 		nlmsvc_unlink_block(block);
 		nlmsvc_release_block(block);
 		goto restart;
 	}
+	spin_unlock(&nlm_blocked_lock);
 	mutex_unlock(&file->f_mutex);
 }
 
+static struct nlm_lockowner *
+nlmsvc_get_lockowner(struct nlm_lockowner *lockowner)
+{
+	refcount_inc(&lockowner->count);
+	return lockowner;
+}
+
+void nlmsvc_put_lockowner(struct nlm_lockowner *lockowner)
+{
+	if (!refcount_dec_and_lock(&lockowner->count, &lockowner->host->h_lock))
+		return;
+	list_del(&lockowner->list);
+	spin_unlock(&lockowner->host->h_lock);
+	nlmsvc_release_host(lockowner->host);
+	kfree(lockowner);
+}
+
+static struct nlm_lockowner *__nlmsvc_find_lockowner(struct nlm_host *host, pid_t pid)
+{
+	struct nlm_lockowner *lockowner;
+	list_for_each_entry(lockowner, &host->h_lockowners, list) {
+		if (lockowner->pid != pid)
+			continue;
+		return nlmsvc_get_lockowner(lockowner);
+	}
+	return NULL;
+}
+
+static struct nlm_lockowner *nlmsvc_find_lockowner(struct nlm_host *host, pid_t pid)
+{
+	struct nlm_lockowner *res, *new = NULL;
+
+	spin_lock(&host->h_lock);
+	res = __nlmsvc_find_lockowner(host, pid);
+
+	if (res == NULL) {
+		spin_unlock(&host->h_lock);
+		new = kmalloc(sizeof(*res), GFP_KERNEL);
+		spin_lock(&host->h_lock);
+		res = __nlmsvc_find_lockowner(host, pid);
+		if (res == NULL && new != NULL) {
+			res = new;
+			/* fs/locks.c will manage the refcount through lock_ops */
+			refcount_set(&new->count, 1);
+			new->pid = pid;
+			new->host = nlm_get_host(host);
+			list_add(&new->list, &host->h_lockowners);
+			new = NULL;
+		}
+	}
+
+	spin_unlock(&host->h_lock);
+	kfree(new);
+	return res;
+}
+
+void
+nlmsvc_release_lockowner(struct nlm_lock *lock)
+{
+	if (lock->fl.c.flc_owner)
+		nlmsvc_put_lockowner(lock->fl.c.flc_owner);
+}
+
+void nlmsvc_locks_init_private(struct file_lock *fl, struct nlm_host *host,
+						pid_t pid)
+{
+	fl->c.flc_owner = nlmsvc_find_lockowner(host, pid);
+}
+
 /*
  * Initialize arguments for GRANTED call. The nlm_rqst structure
  * has been cleared already.
@@ -345,7 +425,7 @@ static int nlmsvc_setgrantargs(struct nlm_rqst *call, struct nlm_lock *lock)
 
 	/* set default data area */
 	call->a_args.lock.oh.data = call->a_owner;
-	call->a_args.lock.svid = lock->fl.fl_pid;
+	call->a_args.lock.svid = ((struct nlm_lockowner *) lock->fl.c.flc_owner)->pid;
 
 	if (lock->oh.len > NLMCLNT_OHSIZE) {
 		void *data = kmalloc(lock->oh.len, GFP_KERNEL);
@@ -400,18 +480,29 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 	    struct nlm_host *host, struct nlm_lock *lock, int wait,
 	    struct nlm_cookie *cookie, int reclaim)
 {
+	struct inode		*inode __maybe_unused = nlmsvc_file_inode(file);
 	struct nlm_block	*block = NULL;
 	int			error;
+	int			mode;
+	int			async_block = 0;
 	__be32			ret;
 
 	dprintk("lockd: nlmsvc_lock(%s/%ld, ty=%d, pi=%d, %Ld-%Ld, bl=%d)\n",
-				locks_inode(file->f_file)->i_sb->s_id,
-				locks_inode(file->f_file)->i_ino,
-				lock->fl.fl_type, lock->fl.fl_pid,
+				inode->i_sb->s_id, inode->i_ino,
+				lock->fl.c.flc_type,
+				lock->fl.c.flc_pid,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end,
 				wait);
 
+	if (nlmsvc_file_cannot_lock(file))
+		return nlm_lck_denied_nolocks;
+
+	if (!locks_can_async_lock(nlmsvc_file_file(file)->f_op)) {
+		async_block = wait;
+		wait = 0;
+	}
+
 	/* Lock file against concurrent access */
 	mutex_lock(&file->f_mutex);
 	/* Get existing block (in case client is busy-waiting)
@@ -425,7 +516,7 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 			goto out;
 		lock = &block->b_call->a_args.lock;
 	} else
-		lock->fl.fl_flags &= ~FL_SLEEP;
+		lock->fl.c.flc_flags &= ~FL_SLEEP;
 
 	if (block->b_flags & B_QUEUED) {
 		dprintk("lockd: nlmsvc_lock deferred block %p flags %d\n",
@@ -453,25 +544,41 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 		goto out;
 	}
 
+	spin_lock(&nlm_blocked_lock);
+	/*
+	 * If this is a lock request for an already pending
+	 * lock request we return nlm_lck_blocked without calling
+	 * vfs_lock_file() again. Otherwise we have two pending
+	 * requests on the underlaying ->lock() implementation but
+	 * only one nlm_block to being granted by lm_grant().
+	 */
+	if (locks_can_async_lock(nlmsvc_file_file(file)->f_op) &&
+	    !list_empty(&block->b_list)) {
+		spin_unlock(&nlm_blocked_lock);
+		ret = nlm_lck_blocked;
+		goto out;
+	}
+
+	/* Append to list of blocked */
+	nlmsvc_insert_block_locked(block, NLM_NEVER);
+	spin_unlock(&nlm_blocked_lock);
+
 	if (!wait)
-		lock->fl.fl_flags &= ~FL_SLEEP;
-	error = vfs_lock_file(file->f_file, F_SETLK, &lock->fl, NULL);
-	lock->fl.fl_flags &= ~FL_SLEEP;
+		lock->fl.c.flc_flags &= ~FL_SLEEP;
+	mode = lock_to_openmode(&lock->fl);
+	error = vfs_lock_file(file->f_file[mode], F_SETLK, &lock->fl, NULL);
+	lock->fl.c.flc_flags &= ~FL_SLEEP;
 
 	dprintk("lockd: vfs_lock_file returned %d\n", error);
 	switch (error) {
 		case 0:
+			nlmsvc_remove_block(block);
 			ret = nlm_granted;
 			goto out;
 		case -EAGAIN:
-			/*
-			 * If this is a blocking request for an
-			 * already pending lock request then we need
-			 * to put it back on lockd's block list
-			 */
-			if (wait)
-				break;
-			ret = nlm_lck_denied;
+			if (!wait)
+				nlmsvc_remove_block(block);
+			ret = async_block ? nlm_lck_blocked : nlm_lck_denied;
 			goto out;
 		case FILE_LOCK_DEFERRED:
 			if (wait)
@@ -481,17 +588,16 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 			ret = nlmsvc_defer_lock_rqst(rqstp, block);
 			goto out;
 		case -EDEADLK:
+			nlmsvc_remove_block(block);
 			ret = nlm_deadlock;
 			goto out;
 		default:			/* includes ENOLCK */
+			nlmsvc_remove_block(block);
 			ret = nlm_lck_denied_nolocks;
 			goto out;
 	}
 
 	ret = nlm_lck_blocked;
-
-	/* Append to list of blocked */
-	nlmsvc_insert_block(block, NLM_NEVER);
 out:
 	mutex_unlock(&file->f_mutex);
 	nlmsvc_release_block(block);
@@ -505,24 +611,29 @@ out:
 __be32
 nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file,
 		struct nlm_host *host, struct nlm_lock *lock,
-		struct nlm_lock *conflock, struct nlm_cookie *cookie)
+		struct nlm_lock *conflock)
 {
 	int			error;
+	int			mode;
 	__be32			ret;
 
 	dprintk("lockd: nlmsvc_testlock(%s/%ld, ty=%d, %Ld-%Ld)\n",
-				locks_inode(file->f_file)->i_sb->s_id,
-				locks_inode(file->f_file)->i_ino,
-				lock->fl.fl_type,
+				nlmsvc_file_inode(file)->i_sb->s_id,
+				nlmsvc_file_inode(file)->i_ino,
+				lock->fl.c.flc_type,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end);
 
+	if (nlmsvc_file_cannot_lock(file))
+		return nlm_lck_denied_nolocks;
+
 	if (locks_in_grace(SVC_NET(rqstp))) {
 		ret = nlm_lck_denied_grace_period;
 		goto out;
 	}
 
-	error = vfs_test_lock(file->f_file, &lock->fl);
+	mode = lock_to_openmode(&lock->fl);
+	error = vfs_test_lock(file->f_file[mode], &lock->fl);
 	if (error) {
 		/* We can't currently deal with deferred test requests */
 		if (error == FILE_LOCK_DEFERRED)
@@ -532,22 +643,23 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file,
 		goto out;
 	}
 
-	if (lock->fl.fl_type == F_UNLCK) {
+	if (lock->fl.c.flc_type == F_UNLCK) {
 		ret = nlm_granted;
 		goto out;
 	}
 
 	dprintk("lockd: conflicting lock(ty=%d, %Ld-%Ld)\n",
-		lock->fl.fl_type, (long long)lock->fl.fl_start,
+		lock->fl.c.flc_type, (long long)lock->fl.fl_start,
 		(long long)lock->fl.fl_end);
 	conflock->caller = "somehost";	/* FIXME */
 	conflock->len = strlen(conflock->caller);
 	conflock->oh.len = 0;		/* don't return OH info */
-	conflock->svid = lock->fl.fl_pid;
-	conflock->fl.fl_type = lock->fl.fl_type;
+	conflock->svid = lock->fl.c.flc_pid;
+	conflock->fl.c.flc_type = lock->fl.c.flc_type;
 	conflock->fl.fl_start = lock->fl.fl_start;
 	conflock->fl.fl_end = lock->fl.fl_end;
 	locks_release_private(&lock->fl);
+
 	ret = nlm_lck_denied;
 out:
 	return ret;
@@ -563,20 +675,30 @@ out:
 __be32
 nlmsvc_unlock(struct net *net, struct nlm_file *file, struct nlm_lock *lock)
 {
-	int	error;
+	int	error = 0;
 
 	dprintk("lockd: nlmsvc_unlock(%s/%ld, pi=%d, %Ld-%Ld)\n",
-				locks_inode(file->f_file)->i_sb->s_id,
-				locks_inode(file->f_file)->i_ino,
-				lock->fl.fl_pid,
+				nlmsvc_file_inode(file)->i_sb->s_id,
+				nlmsvc_file_inode(file)->i_ino,
+				lock->fl.c.flc_pid,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end);
 
+	if (nlmsvc_file_cannot_lock(file))
+		return nlm_lck_denied_nolocks;
+
 	/* First, cancel any lock that might be there */
 	nlmsvc_cancel_blocked(net, file, lock);
 
-	lock->fl.fl_type = F_UNLCK;
-	error = vfs_lock_file(file->f_file, F_SETLK, &lock->fl, NULL);
+	lock->fl.c.flc_type = F_UNLCK;
+	lock->fl.c.flc_file = file->f_file[O_RDONLY];
+	if (lock->fl.c.flc_file)
+		error = vfs_lock_file(lock->fl.c.flc_file, F_SETLK,
+					&lock->fl, NULL);
+	lock->fl.c.flc_file = file->f_file[O_WRONLY];
+	if (lock->fl.c.flc_file)
+		error |= vfs_lock_file(lock->fl.c.flc_file, F_SETLK,
+					&lock->fl, NULL);
 
 	return (error < 0)? nlm_lck_denied_nolocks : nlm_granted;
 }
@@ -593,14 +715,18 @@ nlmsvc_cancel_blocked(struct net *net, struct nlm_file *file, struct nlm_lock *l
 {
 	struct nlm_block	*block;
 	int status = 0;
+	int mode;
 
 	dprintk("lockd: nlmsvc_cancel(%s/%ld, pi=%d, %Ld-%Ld)\n",
-				locks_inode(file->f_file)->i_sb->s_id,
-				locks_inode(file->f_file)->i_ino,
-				lock->fl.fl_pid,
+				nlmsvc_file_inode(file)->i_sb->s_id,
+				nlmsvc_file_inode(file)->i_ino,
+				lock->fl.c.flc_pid,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end);
 
+	if (nlmsvc_file_cannot_lock(file))
+		return nlm_lck_denied_nolocks;
+
 	if (locks_in_grace(net))
 		return nlm_lck_denied_grace_period;
 
@@ -608,8 +734,10 @@ nlmsvc_cancel_blocked(struct net *net, struct nlm_file *file, struct nlm_lock *l
 	block = nlmsvc_lookup_block(file, lock);
 	mutex_unlock(&file->f_mutex);
 	if (block != NULL) {
-		vfs_cancel_lock(block->b_file->f_file,
-				&block->b_call->a_args.lock.fl);
+		struct file_lock *fl = &block->b_call->a_args.lock.fl;
+
+		mode = lock_to_openmode(fl);
+		vfs_cancel_lock(block->b_file->f_file[mode], fl);
 		status = nlmsvc_unlink_block(block);
 		nlmsvc_release_block(block);
 	}
@@ -692,27 +820,21 @@ nlmsvc_notify_blocked(struct file_lock *fl)
 	printk(KERN_WARNING "lockd: notification for unknown block!\n");
 }
 
-static int nlmsvc_same_owner(struct file_lock *fl1, struct file_lock *fl2)
+static fl_owner_t nlmsvc_get_owner(fl_owner_t owner)
 {
-	return fl1->fl_owner == fl2->fl_owner && fl1->fl_pid == fl2->fl_pid;
+	return nlmsvc_get_lockowner(owner);
 }
 
-/*
- * Since NLM uses two "keys" for tracking locks, we need to hash them down
- * to one for the blocked_hash. Here, we're just xor'ing the host address
- * with the pid in order to create a key value for picking a hash bucket.
- */
-static unsigned long
-nlmsvc_owner_key(struct file_lock *fl)
+static void nlmsvc_put_owner(fl_owner_t owner)
 {
-	return (unsigned long)fl->fl_owner ^ (unsigned long)fl->fl_pid;
+	nlmsvc_put_lockowner(owner);
 }
 
 const struct lock_manager_operations nlmsvc_lock_operations = {
-	.lm_compare_owner = nlmsvc_same_owner,
-	.lm_owner_key = nlmsvc_owner_key,
 	.lm_notify = nlmsvc_notify_blocked,
 	.lm_grant = nlmsvc_grant_deferred,
+	.lm_get_owner = nlmsvc_get_owner,
+	.lm_put_owner = nlmsvc_put_owner,
 };
 
 /*
@@ -731,6 +853,7 @@ nlmsvc_grant_blocked(struct nlm_block *block)
 {
 	struct nlm_file		*file = block->b_file;
 	struct nlm_lock		*lock = &block->b_call->a_args.lock;
+	int			mode;
 	int			error;
 	loff_t			fl_start, fl_end;
 
@@ -753,11 +876,12 @@ nlmsvc_grant_blocked(struct nlm_block *block)
 	/* vfs_lock_file() can mangle fl_start and fl_end, but we need
 	 * them unchanged for the GRANT_MSG
 	 */
-	lock->fl.fl_flags |= FL_SLEEP;
+	lock->fl.c.flc_flags |= FL_SLEEP;
 	fl_start = lock->fl.fl_start;
 	fl_end = lock->fl.fl_end;
-	error = vfs_lock_file(file->f_file, F_SETLK, &lock->fl, NULL);
-	lock->fl.fl_flags &= ~FL_SLEEP;
+	mode = lock_to_openmode(&lock->fl);
+	error = vfs_lock_file(file->f_file[mode], F_SETLK, &lock->fl, NULL);
+	lock->fl.c.flc_flags &= ~FL_SLEEP;
 	lock->fl.fl_start = fl_start;
 	lock->fl.fl_end = fl_end;
 
@@ -865,19 +989,32 @@ void
 nlmsvc_grant_reply(struct nlm_cookie *cookie, __be32 status)
 {
 	struct nlm_block	*block;
+	struct file_lock	*fl;
+	int			error;
 
-	dprintk("grant_reply: looking for cookie %x, s=%d \n",
+	dprintk("grant_reply: looking for cookie %x, s=%d\n",
 		*(unsigned int *)(cookie->data), status);
 	if (!(block = nlmsvc_find_block(cookie)))
 		return;
 
-	if (status == nlm_lck_denied_grace_period) {
+	switch (status) {
+	case nlm_lck_denied_grace_period:
 		/* Try again in a couple of seconds */
 		nlmsvc_insert_block(block, 10 * HZ);
-	} else {
+		break;
+	case nlm_lck_denied:
+		/* Client doesn't want it, just unlock it */
+		nlmsvc_unlink_block(block);
+		fl = &block->b_call->a_args.lock.fl;
+		fl->c.flc_type = F_UNLCK;
+		error = vfs_lock_file(fl->c.flc_file, F_SETLK, fl, NULL);
+		if (error)
+			pr_warn("lockd: unable to unlock lock rejected by client!\n");
+		break;
+	default:
 		/*
-		 * Lock is now held by client, or has been rejected.
-		 * In both cases, the block should be removed.
+		 * Either it was accepted or the status makes no sense
+		 * just unlink it either way.
 		 */
 		nlmsvc_unlink_block(block);
 	}
@@ -906,14 +1043,14 @@ retry_deferred_block(struct nlm_block *block)
  * picks up locks that can be granted, or grant notifications that must
  * be retransmitted.
  */
-unsigned long
-nlmsvc_retry_blocked(void)
+void
+nlmsvc_retry_blocked(struct svc_rqst *rqstp)
 {
 	unsigned long	timeout = MAX_SCHEDULE_TIMEOUT;
 	struct nlm_block *block;
 
 	spin_lock(&nlm_blocked_lock);
-	while (!list_empty(&nlm_blocked) && !kthread_should_stop()) {
+	while (!list_empty(&nlm_blocked) && !svc_thread_should_stop(rqstp)) {
 		block = list_entry(nlm_blocked.next, struct nlm_block, b_list);
 
 		if (block->b_when == NLM_NEVER)
@@ -936,5 +1073,6 @@ nlmsvc_retry_blocked(void)
 	}
 	spin_unlock(&nlm_blocked_lock);
 
-	return timeout;
+	if (timeout < MAX_SCHEDULE_TIMEOUT)
+		mod_timer(&nlmsvc_retry, jiffies + timeout);
 }