dlm: make plock operation killable

Allow processes blocked on plock requests to be interrupted
when they are killed.  This leaves the problem of cleaning
up the lock state in userspace.  This has three parts:

1. Add a flag to unlock operations sent to userspace
indicating the file is being closed.  Userspace will
then look for and clear any waiting plock operations that
were abandoned by an interrupted process.

2. Queue an unlock-close operation (like in 1) to clean up
userspace from an interrupted plock request.  This is needed
because the vfs will not send a cleanup-unlock if it sees no
locks on the file, which it won't if the interrupted operation
was the only one.

3. Do not use replies from userspace for unlock-close operations
because they are unnecessary (they are just cleaning up for the
process which did not make an unlock call).  This also simplifies
the new unlock-close generated from point 2.

Signed-off-by: David Teigland <teigland@redhat.com>

fs/dlm/plock.c

@@ -71,6 +71,36 @@ static void send_op(struct plock_op *op)
 	wake_up(&send_wq);
 }
 
+/* If a process was killed while waiting for the only plock on a file,
+   locks_remove_posix will not see any lock on the file so it won't
+   send an unlock-close to us to pass on to userspace to clean up the
+   abandoned waiter.  So, we have to insert the unlock-close when the
+   lock call is interrupted. */
+
+static void do_unlock_close(struct dlm_ls *ls, u64 number,
+			    struct file *file, struct file_lock *fl)
+{
+	struct plock_op *op;
+
+	op = kzalloc(sizeof(*op), GFP_NOFS);
+	if (!op)
+		return;
+
+	op->info.optype		= DLM_PLOCK_OP_UNLOCK;
+	op->info.pid		= fl->fl_pid;
+	op->info.fsid		= ls->ls_global_id;
+	op->info.number		= number;
+	op->info.start		= 0;
+	op->info.end		= OFFSET_MAX;
+	if (fl->fl_lmops && fl->fl_lmops->fl_grant)
+		op->info.owner	= (__u64) fl->fl_pid;
+	else
+		op->info.owner	= (__u64)(long) fl->fl_owner;
+
+	op->info.flags |= DLM_PLOCK_FL_CLOSE;
+	send_op(op);
+}
+
 int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 		   int cmd, struct file_lock *fl)
 {
@@ -114,9 +144,19 @@ int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 
 	send_op(op);
 
-	if (xop->callback == NULL)
-		wait_event(recv_wq, (op->done != 0));
-	else {
+	if (xop->callback == NULL) {
+		rv = wait_event_killable(recv_wq, (op->done != 0));
+		if (rv == -ERESTARTSYS) {
+			log_debug(ls, "dlm_posix_lock: wait killed %llx",
+				  (unsigned long long)number);
+			spin_lock(&ops_lock);
+			list_del(&op->list);
+			spin_unlock(&ops_lock);
+			kfree(xop);
+			do_unlock_close(ls, number, file, fl);
+			goto out;
+		}
+	} else {
 		rv = FILE_LOCK_DEFERRED;
 		goto out;
 	}
@@ -233,6 +273,13 @@ int dlm_posix_unlock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	else
 		op->info.owner	= (__u64)(long) fl->fl_owner;
 
+	if (fl->fl_flags & FL_CLOSE) {
+		op->info.flags |= DLM_PLOCK_FL_CLOSE;
+		send_op(op);
+		rv = 0;
+		goto out;
+	}
+
 	send_op(op);
 	wait_event(recv_wq, (op->done != 0));
 
@@ -334,14 +381,24 @@ static ssize_t dev_read(struct file *file, char __user *u, size_t count,
 	spin_lock(&ops_lock);
 	if (!list_empty(&send_list)) {
 		op = list_entry(send_list.next, struct plock_op, list);
-		list_move(&op->list, &recv_list);
+		if (op->info.flags & DLM_PLOCK_FL_CLOSE)
+			list_del(&op->list);
+		else
+			list_move(&op->list, &recv_list);
 		memcpy(&info, &op->info, sizeof(info));
 	}
 	spin_unlock(&ops_lock);
 
 	if (!op)
 		return -EAGAIN;
 
+	/* there is no need to get a reply from userspace for unlocks
+	   that were generated by the vfs cleaning up for a close
+	   (the process did not make an unlock call). */
+
+	if (op->info.flags & DLM_PLOCK_FL_CLOSE)
+		kfree(op);
+
 	if (copy_to_user(u, &info, sizeof(info)))
 		return -EFAULT;
 	return sizeof(info);

include/linux/dlm_plock.h

@@ -14,7 +14,7 @@
 #define DLM_PLOCK_MISC_NAME		"dlm_plock"
 
 #define DLM_PLOCK_VERSION_MAJOR	1
-#define DLM_PLOCK_VERSION_MINOR	1
+#define DLM_PLOCK_VERSION_MINOR	2
 #define DLM_PLOCK_VERSION_PATCH	0
 
 enum {
@@ -23,12 +23,14 @@ enum {
 	DLM_PLOCK_OP_GET,
 };
 
+#define DLM_PLOCK_FL_CLOSE 1
+
 struct dlm_plock_info {
 	__u32 version[3];
 	__u8 optype;
 	__u8 ex;
 	__u8 wait;
-	__u8 pad;
+	__u8 flags;
 	__u32 pid;
 	__s32 nodeid;
 	__s32 rv;