Merge branch 'locks' of git://linux-nfs.org/~bfields/linux

* 'locks' of git://linux-nfs.org/~bfields/linux:
  nfsd: remove IS_ISMNDLCK macro
  Rework /proc/locks via seq_files and seq_list helpers
  fs/locks.c: use list_for_each_entry() instead of list_for_each()
  NFS: clean up explicit check for mandatory locks
  AFS: clean up explicit check for mandatory locks
  9PFS: clean up explicit check for mandatory locks
  GFS2: clean up explicit check for mandatory locks
  Cleanup macros for distinguishing mandatory locks
  Documentation: move locks.txt in filesystems/
  locks: add warning about mandatory locking races
  Documentation: move mandatory locking documentation to filesystems/
  locks: Fix potential OOPS in generic_setlease()
  Use list_first_entry in locks_wake_up_blocks
  locks: fix flock_lock_file() comment
  Memory shortage can result in inconsistent flocks state
  locks: kill redundant local variable
  locks: reverse order of posix_locks_conflict() arguments

Documentation/00-INDEX

@@ -145,7 +145,7 @@ fb/
 feature-removal-schedule.txt
 	- list of files and features that are going to be removed.
 filesystems/
-	- directory with info on the various filesystems that Linux supports.
+	- info on the vfs and the various filesystems that Linux supports.
 firmware_class/
 	- request_firmware() hotplug interface info.
 floppy.txt
@@ -230,8 +230,6 @@ local_ops.txt
 	- semantics and behavior of local atomic operations.
 lockdep-design.txt
 	- documentation on the runtime locking correctness validator.
-locks.txt
-	- info on file locking implementations, flock() vs. fcntl(), etc.
 logo.gif
 	- full colour GIF image of Linux logo (penguin - Tux).
 logo.txt
@@ -240,8 +238,6 @@ m68k/
 	- directory with info about Linux on Motorola 68k architecture.
 magic-number.txt
 	- list of magic numbers used to mark/protect kernel data structures.
-mandatory.txt
-	- info on the Linux implementation of Sys V mandatory file locking.
 mca.txt
 	- info on supporting Micro Channel Architecture (e.g. PS/2) systems.
 md.txt

Documentation/filesystems/00-INDEX

@@ -52,6 +52,10 @@ isofs.txt
 	- info and mount options for the ISO 9660 (CDROM) filesystem.
 jfs.txt
 	- info and mount options for the JFS filesystem.
+locks.txt
+	- info on file locking implementations, flock() vs. fcntl(), etc.
+mandatory-locking.txt
+	- info on the Linux implementation of Sys V mandatory file locking.
 ncpfs.txt
 	- info on Novell Netware(tm) filesystem using NCP protocol.
 ntfs.txt

Documentation/locks.txt → Documentation/filesystems/locks.txt

@@ -53,11 +53,11 @@ fcntl(), with all the problems that implies.
 1.3 Mandatory Locking As A Mount Option
 ---------------------------------------
 
-Mandatory locking, as described in 'Documentation/mandatory.txt' was prior
-to this release a general configuration option that was valid for all
-mounted filesystems. This had a number of inherent dangers, not the least
-of which was the ability to freeze an NFS server by asking it to read a
-file for which a mandatory lock existed.
+Mandatory locking, as described in 'Documentation/filesystems/mandatory.txt'
+was prior to this release a general configuration option that was valid for
+all mounted filesystems.  This had a number of inherent dangers, not the
+least of which was the ability to freeze an NFS server by asking it to read
+a file for which a mandatory lock existed.
 
 From this release of the kernel, mandatory locking can be turned on and off
 on a per-filesystem basis, using the mount options 'mand' and 'nomand'.

Documentation/mandatory.txt → Documentation/filesystems/mandatory-locking.txt

@@ -3,7 +3,26 @@
 		Andy Walker <andy@lysaker.kvaerner.no>
 
 			   15 April 1996
-
+		     (Updated September 2007)
+
+0. Why you should avoid mandatory locking
+-----------------------------------------
+
+The Linux implementation is prey to a number of difficult-to-fix race
+conditions which in practice make it not dependable:
+
+	- The write system call checks for a mandatory lock only once
+	  at its start.  It is therefore possible for a lock request to
+	  be granted after this check but before the data is modified.
+	  A process may then see file data change even while a mandatory
+	  lock was held.
+	- Similarly, an exclusive lock may be granted on a file after
+	  the kernel has decided to proceed with a read, but before the
+	  read has actually completed, and the reading process may see
+	  the file data in a state which should not have been visible
+	  to it.
+	- Similar races make the claimed mutual exclusion between lock
+	  and mmap similarly unreliable.
 
 1. What is  mandatory locking?
 ------------------------------

fs/9p/vfs_file.c

@@ -105,7 +105,7 @@ static int v9fs_file_lock(struct file *filp, int cmd, struct file_lock *fl)
 	P9_DPRINTK(P9_DEBUG_VFS, "filp: %p lock: %p\n", filp, fl);
 
 	/* No mandatory locks */
-	if ((inode->i_mode & (S_ISGID | S_IXGRP)) == S_ISGID)
+	if (__mandatory_lock(inode))
 		return -ENOLCK;
 
 	if ((IS_SETLK(cmd) || IS_SETLKW(cmd)) && fl->fl_type != F_UNLCK) {

fs/afs/flock.c

@@ -524,8 +524,7 @@ int afs_lock(struct file *file, int cmd, struct file_lock *fl)
 	       (long long) fl->fl_start, (long long) fl->fl_end);
 
 	/* AFS doesn't support mandatory locks */
-	if ((vnode->vfs_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID &&
-	    fl->fl_type != F_UNLCK)
+	if (__mandatory_lock(&vnode->vfs_inode) && fl->fl_type != F_UNLCK)
 		return -ENOLCK;
 
 	if (IS_GETLK(cmd))

fs/gfs2/ops_file.c

@@ -535,7 +535,7 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl)
 
 	if (!(fl->fl_flags & FL_POSIX))
 		return -ENOLCK;
-	if ((ip->i_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID)
+	if (__mandatory_lock(&ip->i_inode))
 		return -ENOLCK;
 
 	if (sdp->sd_args.ar_localflocks) {
@@ -636,7 +636,7 @@ static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl)
 
 	if (!(fl->fl_flags & FL_FLOCK))
 		return -ENOLCK;
-	if ((ip->i_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID)
+	if (__mandatory_lock(&ip->i_inode))
 		return -ENOLCK;
 
 	if (sdp->sd_args.ar_localflocks)

fs/nfs/file.c

@@ -577,8 +577,7 @@ static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl)
 	nfs_inc_stats(inode, NFSIOS_VFSLOCK);
 
 	/* No mandatory locks over NFS */
-	if ((inode->i_mode & (S_ISGID | S_IXGRP)) == S_ISGID &&
-	    fl->fl_type != F_UNLCK)
+	if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK)
 		return -ENOLCK;
 
 	if (IS_GETLK(cmd))

fs/nfsd/nfs4state.c

@@ -2035,7 +2035,7 @@ static inline int
 io_during_grace_disallowed(struct inode *inode, int flags)
 {
 	return nfs4_in_grace() && (flags & (RD_STATE | WR_STATE))
-		&& MANDATORY_LOCK(inode);
+		&& mandatory_lock(inode);
 }
 
 /*

fs/nfsd/vfs.c

@@ -61,12 +61,6 @@
 #define NFSDDBG_FACILITY		NFSDDBG_FILEOP
 
 
-/* We must ignore files (but only files) which might have mandatory
- * locks on them because there is no way to know if the accesser has
- * the lock.
- */
-#define IS_ISMNDLK(i)	(S_ISREG((i)->i_mode) && MANDATORY_LOCK(i))
-
 /*
  * This is a cache of readahead params that help us choose the proper
  * readahead strategy. Initially, we set all readahead parameters to 0
@@ -689,7 +683,12 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	err = nfserr_perm;
 	if (IS_APPEND(inode) && (access & MAY_WRITE))
 		goto out;
-	if (IS_ISMNDLK(inode))
+	/*
+	 * We must ignore files (but only files) which might have mandatory
+	 * locks on them because there is no way to know if the accesser has
+	 * the lock.
+	 */
+	if (S_ISREG((inode)->i_mode) && mandatory_lock(inode))
 		goto out;
 
 	if (!inode->i_fop)

fs/proc/proc_misc.c

@@ -66,7 +66,6 @@ extern int get_stram_list(char *);
 extern int get_filesystem_list(char *);
 extern int get_exec_domain_list(char *);
 extern int get_dma_list(char *);
-extern int get_locks_status (char *, char **, off_t, int);
 
 static int proc_calc_metrics(char *page, char **start, off_t off,
 				 int count, int *eof, int len)
@@ -624,16 +623,18 @@ static int cmdline_read_proc(char *page, char **start, off_t off,
 	return proc_calc_metrics(page, start, off, count, eof, len);
 }
 
-static int locks_read_proc(char *page, char **start, off_t off,
-				 int count, int *eof, void *data)
+static int locks_open(struct inode *inode, struct file *filp)
 {
-	int len = get_locks_status(page, start, off, count);
-
-	if (len < count)
-		*eof = 1;
-	return len;
+	return seq_open(filp, &locks_seq_operations);
 }
 
+static const struct file_operations proc_locks_operations = {
+	.open		= locks_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
 static int execdomains_read_proc(char *page, char **start, off_t off,
 				 int count, int *eof, void *data)
 {
@@ -691,7 +692,6 @@ void __init proc_misc_init(void)
 #endif
 		{"filesystems",	filesystems_read_proc},
 		{"cmdline",	cmdline_read_proc},
-		{"locks",	locks_read_proc},
 		{"execdomains",	execdomains_read_proc},
 		{NULL,}
 	};
@@ -709,6 +709,7 @@ void __init proc_misc_init(void)
 			entry->proc_fops = &proc_kmsg_operations;
 	}
 #endif
+	create_seq_entry("locks", 0, &proc_locks_operations);
 	create_seq_entry("devices", 0, &proc_devinfo_operations);
 	create_seq_entry("cpuinfo", 0, &proc_cpuinfo_operations);
 #ifdef CONFIG_BLOCK

fs/read_write.c

@@ -205,7 +205,7 @@ int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count
 	if (unlikely((pos < 0) || (loff_t) (pos + count) < 0))
 		goto Einval;
 
-	if (unlikely(inode->i_flock && MANDATORY_LOCK(inode))) {
+	if (unlikely(inode->i_flock && mandatory_lock(inode))) {
 		int retval = locks_mandatory_area(
 			read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE,
 			inode, file, pos, count);

include/linux/fs.h

@@ -883,6 +883,7 @@ extern int vfs_setlease(struct file *, long, struct file_lock **);
 extern int lease_modify(struct file_lock **, int);
 extern int lock_may_read(struct inode *, loff_t start, unsigned long count);
 extern int lock_may_write(struct inode *, loff_t start, unsigned long count);
+extern struct seq_operations locks_seq_operations;
 
 struct fasync_struct {
 	int	magic;
@@ -1375,12 +1376,25 @@ extern int locks_mandatory_area(int, struct inode *, struct file *, loff_t, size
  * Candidates for mandatory locking have the setgid bit set
  * but no group execute bit -  an otherwise meaningless combination.
  */
-#define MANDATORY_LOCK(inode) \
-	(IS_MANDLOCK(inode) && ((inode)->i_mode & (S_ISGID | S_IXGRP)) == S_ISGID)
+
+static inline int __mandatory_lock(struct inode *ino)
+{
+	return (ino->i_mode & (S_ISGID | S_IXGRP)) == S_ISGID;
+}
+
+/*
+ * ... and these candidates should be on MS_MANDLOCK mounted fs,
+ * otherwise these will be advisory locks
+ */
+
+static inline int mandatory_lock(struct inode *ino)
+{
+	return IS_MANDLOCK(ino) && __mandatory_lock(ino);
+}
 
 static inline int locks_verify_locked(struct inode *inode)
 {
-	if (MANDATORY_LOCK(inode))
+	if (mandatory_lock(inode))
 		return locks_mandatory_locked(inode);
 	return 0;
 }
@@ -1391,7 +1405,7 @@ static inline int locks_verify_truncate(struct inode *inode,
 				    struct file *filp,
 				    loff_t size)
 {
-	if (inode->i_flock && MANDATORY_LOCK(inode))
+	if (inode->i_flock && mandatory_lock(inode))
 		return locks_mandatory_area(
 			FLOCK_VERIFY_WRITE, inode, filp,
 			size < inode->i_size ? size : inode->i_size,