Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  ceph: fix ioctl magic
  ceph: Behave better when handling file lock replies.
  ceph: pass lock information by struct file_lock instead of as individual params.
  ceph: Handle file locks in replies from the MDS.
  ceph: avoid possible null deref in readdir after dir llseek

5 files changed, 111 insertions, 61 deletions

diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 7d447af84ec4..158c700fdca5 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -114,8 +114,8 @@ static int __dcache_readdir(struct file *filp,
 	spin_lock(&dcache_lock);
 
 	/* start at beginning? */
-	if (filp->f_pos == 2 || (last &&
-				 filp->f_pos < ceph_dentry(last)->offset)) {
+	if (filp->f_pos == 2 || last == NULL ||
+	    filp->f_pos < ceph_dentry(last)->offset) {
 		if (list_empty(&parent->d_subdirs))
 			goto out_unlock;
 		p = parent->d_subdirs.prev;
diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h
index a6ce54e94eb5..52e8fd74d450 100644
--- a/fs/ceph/ioctl.h
+++ b/fs/ceph/ioctl.h
@@ -4,7 +4,7 @@
 #include <linux/ioctl.h>
 #include <linux/types.h>
 
-#define CEPH_IOCTL_MAGIC 0x98
+#define CEPH_IOCTL_MAGIC 0x97
 
 /* just use u64 to align sanely on all archs */
 struct ceph_ioctl_layout {
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 40abde93c345..476b329867d4 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -11,40 +11,68 @@
  * Implement fcntl and flock locking functions.
  */
 static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
-			     u64 pid, u64 pid_ns,
-			     int cmd, u64 start, u64 length, u8 wait)
+			     int cmd, u8 wait, struct file_lock *fl)
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_mds_client *mdsc =
 		ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_mds_request *req;
 	int err;
+	u64 length = 0;
 
 	req = ceph_mdsc_create_request(mdsc, operation, USE_AUTH_MDS);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 	req->r_inode = igrab(inode);
 
+	/* mds requires start and length rather than start and end */
+	if (LLONG_MAX == fl->fl_end)
+		length = 0;
+	else
+		length = fl->fl_end - fl->fl_start + 1;
+
 	dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
 	     "length: %llu, wait: %d, type`: %d", (int)lock_type,
-	     (int)operation, pid, start, length, wait, cmd);
+	     (int)operation, (u64)fl->fl_pid, fl->fl_start,
+	     length, wait, fl->fl_type);
+
 
 	req->r_args.filelock_change.rule = lock_type;
 	req->r_args.filelock_change.type = cmd;
-	req->r_args.filelock_change.pid = cpu_to_le64(pid);
+	req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid);
 	/* This should be adjusted, but I'm not sure if
 	   namespaces actually get id numbers*/
 	req->r_args.filelock_change.pid_namespace =
-		cpu_to_le64((u64)pid_ns);
-	req->r_args.filelock_change.start = cpu_to_le64(start);
+		cpu_to_le64((u64)(unsigned long)fl->fl_nspid);
+	req->r_args.filelock_change.start = cpu_to_le64(fl->fl_start);
 	req->r_args.filelock_change.length = cpu_to_le64(length);
 	req->r_args.filelock_change.wait = wait;
 
 	err = ceph_mdsc_do_request(mdsc, inode, req);
+
+	if ( operation == CEPH_MDS_OP_GETFILELOCK){
+		fl->fl_pid = le64_to_cpu(req->r_reply_info.filelock_reply->pid);
+		if (CEPH_LOCK_SHARED == req->r_reply_info.filelock_reply->type)
+			fl->fl_type = F_RDLCK;
+		else if (CEPH_LOCK_EXCL == req->r_reply_info.filelock_reply->type)
+			fl->fl_type = F_WRLCK;
+		else
+			fl->fl_type = F_UNLCK;
+
+		fl->fl_start = le64_to_cpu(req->r_reply_info.filelock_reply->start);
+		length = le64_to_cpu(req->r_reply_info.filelock_reply->start) +
+						 le64_to_cpu(req->r_reply_info.filelock_reply->length);
+		if (length >= 1)
+			fl->fl_end = length -1;
+		else
+			fl->fl_end = 0;
+
+	}
 	ceph_mdsc_put_request(req);
 	dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
-	     "length: %llu, wait: %d, type`: %d err code %d", (int)lock_type,
-	     (int)operation, pid, start, length, wait, cmd, err);
+	     "length: %llu, wait: %d, type`: %d, err code %d", (int)lock_type,
+	     (int)operation, (u64)fl->fl_pid, fl->fl_start,
+	     length, wait, fl->fl_type, err);
 	return err;
 }
 
@@ -54,7 +82,6 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
  */
 int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
 {
-	u64 length;
 	u8 lock_cmd;
 	int err;
 	u8 wait = 0;
@@ -76,29 +103,20 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
 	else
 		lock_cmd = CEPH_LOCK_UNLOCK;
 
-	if (LLONG_MAX == fl->fl_end)
-		length = 0;
-	else
-		length = fl->fl_end - fl->fl_start + 1;
-
-	err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
-				(u64)fl->fl_pid,
-				(u64)(unsigned long)fl->fl_nspid,
-				lock_cmd, fl->fl_start,
-				length, wait);
+	err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file, lock_cmd, wait, fl);
 	if (!err) {
-		dout("mds locked, locking locally");
-		err = posix_lock_file(file, fl, NULL);
-		if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
-			/* undo! This should only happen if the kernel detects
-			 * local deadlock. */
-			ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
-					  (u64)fl->fl_pid,
-					  (u64)(unsigned long)fl->fl_nspid,
-					  CEPH_LOCK_UNLOCK, fl->fl_start,
-					  length, 0);
-			dout("got %d on posix_lock_file, undid lock", err);
+		if ( op != CEPH_MDS_OP_GETFILELOCK ){
+			dout("mds locked, locking locally");
+			err = posix_lock_file(file, fl, NULL);
+			if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
+				/* undo! This should only happen if the kernel detects
+				 * local deadlock. */
+				ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
+						  CEPH_LOCK_UNLOCK, 0, fl);
+				dout("got %d on posix_lock_file, undid lock", err);
+			}
 		}
+
 	} else {
 		dout("mds returned error code %d", err);
 	}
@@ -107,7 +125,6 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
 
 int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
 {
-	u64 length;
 	u8 lock_cmd;
 	int err;
 	u8 wait = 1;
@@ -127,26 +144,15 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
 		lock_cmd = CEPH_LOCK_EXCL;
 	else
 		lock_cmd = CEPH_LOCK_UNLOCK;
-	/* mds requires start and length rather than start and end */
-	if (LLONG_MAX == fl->fl_end)
-		length = 0;
-	else
-		length = fl->fl_end - fl->fl_start + 1;
 
 	err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK,
-				file, (u64)fl->fl_pid,
-				(u64)(unsigned long)fl->fl_nspid,
-				lock_cmd, fl->fl_start,
-				length, wait);
+				file, lock_cmd, wait, fl);
 	if (!err) {
 		err = flock_lock_file_wait(file, fl);
 		if (err) {
 			ceph_lock_message(CEPH_LOCK_FLOCK,
 					  CEPH_MDS_OP_SETFILELOCK,
-					  file, (u64)fl->fl_pid,
-					  (u64)(unsigned long)fl->fl_nspid,
-					  CEPH_LOCK_UNLOCK, fl->fl_start,
-					  length, 0);
+					  file, CEPH_LOCK_UNLOCK, 0, fl);
 			dout("got %d on flock_lock_file_wait, undid lock", err);
 		}
 	} else {
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 098b18508479..38800eaa81d0 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -202,6 +202,38 @@ out_bad:
 }
 
 /*
+ * parse fcntl F_GETLK results
+ */
+static int parse_reply_info_filelock(void **p, void *end,
+                struct ceph_mds_reply_info_parsed *info)
+{
+	if (*p + sizeof(*info->filelock_reply) > end)
+		goto bad;
+
+	info->filelock_reply = *p;
+	*p += sizeof(*info->filelock_reply);
+
+	if (unlikely(*p != end))
+		goto bad;
+	return 0;
+
+bad:
+	return -EIO;
+}
+
+/*
+ * parse extra results
+ */
+static int parse_reply_info_extra(void **p, void *end,
+                struct ceph_mds_reply_info_parsed *info)
+{
+	if (info->head->op == CEPH_MDS_OP_GETFILELOCK)
+		return parse_reply_info_filelock(p, end, info);
+	else
+		return parse_reply_info_dir(p, end, info);
+}
+
+/*
  * parse entire mds reply
  */
 static int parse_reply_info(struct ceph_msg *msg,
@@ -223,10 +255,10 @@ static int parse_reply_info(struct ceph_msg *msg,
 			goto out_bad;
 	}
 
-	/* dir content */
+	/* extra */
 	ceph_decode_32_safe(&p, end, len, bad);
 	if (len > 0) {
-		err = parse_reply_info_dir(&p, p+len, info);
+		err = parse_reply_info_extra(&p, p+len, info);
 		if (err < 0)
 			goto out_bad;
 	}
@@ -2074,7 +2106,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 
 	mutex_lock(&session->s_mutex);
 	if (err < 0) {
-		pr_err("mdsc_handle_reply got corrupt reply mds%d\n", mds);
+		pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid);
 		ceph_msg_dump(msg);
 		goto out_err;
 	}
@@ -2094,7 +2126,8 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 	mutex_lock(&req->r_fill_mutex);
 	err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
 	if (err == 0) {
-		if (result == 0 && rinfo->dir_nr)
+		if (result == 0 && req->r_op != CEPH_MDS_OP_GETFILELOCK &&
+		    rinfo->dir_nr)
 			ceph_readdir_prepopulate(req, req->r_session);
 		ceph_unreserve_caps(mdsc, &req->r_caps_reservation);
 	}
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index 9341fd4f1432..aabe563b54db 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -42,26 +42,37 @@ struct ceph_mds_reply_info_in {
 };
 
 /*
- * parsed info about an mds reply, including information about the
- * target inode and/or its parent directory and dentry, and directory
- * contents (for readdir results).
+ * parsed info about an mds reply, including information about
+ * either: 1) the target inode and/or its parent directory and dentry,
+ * and directory contents (for readdir results), or
+ * 2) the file range lock info (for fcntl F_GETLK results).
  */
 struct ceph_mds_reply_info_parsed {
 	struct ceph_mds_reply_head    *head;
 
+	/* trace */
 	struct ceph_mds_reply_info_in diri, targeti;
 	struct ceph_mds_reply_dirfrag *dirfrag;
 	char                          *dname;
 	u32                           dname_len;
 	struct ceph_mds_reply_lease   *dlease;
 
-	struct ceph_mds_reply_dirfrag *dir_dir;
-	int                           dir_nr;
-	char                          **dir_dname;
-	u32                           *dir_dname_len;
-	struct ceph_mds_reply_lease   **dir_dlease;
-	struct ceph_mds_reply_info_in *dir_in;
-	u8                            dir_complete, dir_end;
+	/* extra */
+	union {
+		/* for fcntl F_GETLK results */
+		struct ceph_filelock *filelock_reply;
+
+		/* for readdir results */
+		struct {
+			struct ceph_mds_reply_dirfrag *dir_dir;
+			int                           dir_nr;
+			char                          **dir_dname;
+			u32                           *dir_dname_len;
+			struct ceph_mds_reply_lease   **dir_dlease;
+			struct ceph_mds_reply_info_in *dir_in;
+			u8                            dir_complete, dir_end;
+		};
+	};
 
 	/* encoded blob describing snapshot contexts for certain
 	   operations (e.g., open) */