[XFS] Make hole punching at EOF atomic.

If hole punching at EOF is done as two steps (i.e. truncate then extend)
the file is in a transient state between the two steps where an
application can see the incorrect file size. Punching a hole to EOF needs
to be treated in teh same way as all other hole punching cases so that the
file size is never seen to change.

SGI-PV: 962012
SGI-Modid: xfs-linux-melb:xfs-kern:28641a

Signed-off-by: David Chinner <dgc@sgi.com>
Signed-off-by: Vlad Apostolov <vapo@sgi.com>
Signed-off-by: Tim Shimmin <tes@sgi.com>

2 files changed, 29 insertions, 13 deletions

diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h
index 188b296..ba74000 100644
--- a/fs/xfs/xfs_rw.h
+++ b/fs/xfs/xfs_rw.h
@@ -72,6 +72,12 @@ xfs_fsb_to_db_io(struct xfs_iocore *io, xfs_fsblock_t fsb)
 }
 
 /*
+ * Flags for xfs_free_eofblocks
+ */
+#define XFS_FREE_EOF_LOCK	(1<<0)
+#define XFS_FREE_EOF_NOLOCK	(1<<1)
+
+/*
  * Prototypes for functions in xfs_rw.c.
  */
 extern int xfs_write_clear_setuid(struct xfs_inode *ip);
@@ -91,10 +97,12 @@ extern void xfs_ioerror_alert(char *func, struct xfs_mount *mp,
 extern int xfs_rwlock(bhv_desc_t *bdp, bhv_vrwlock_t write_lock);
 extern void xfs_rwunlock(bhv_desc_t *bdp, bhv_vrwlock_t write_lock);
 extern int xfs_setattr(bhv_desc_t *, bhv_vattr_t *vap, int flags,
-		       cred_t *credp);
+			cred_t *credp);
 extern int xfs_change_file_space(bhv_desc_t *bdp, int cmd, xfs_flock64_t *bf,
-				 xfs_off_t offset, cred_t *credp, int flags);
+			xfs_off_t offset, cred_t *credp, int flags);
 extern int xfs_set_dmattrs(bhv_desc_t *bdp, u_int evmask, u_int16_t state,
-			   cred_t *credp);
+			cred_t *credp);
+extern int xfs_free_eofblocks(struct xfs_mount *mp, struct xfs_inode *ip,
+			int flags);
 
 #endif /* __XFS_RW_H__ */
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 70bc82f..6420ca8 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -1201,13 +1201,15 @@ xfs_fsync(
 }
 
 /*
- * This is called by xfs_inactive to free any blocks beyond eof,
- * when the link count isn't zero.
+ * This is called by xfs_inactive to free any blocks beyond eof
+ * when the link count isn't zero and by xfs_dm_punch_hole() when
+ * punching a hole to EOF.
  */
-STATIC int
-xfs_inactive_free_eofblocks(
+int
+xfs_free_eofblocks(
 	xfs_mount_t	*mp,
-	xfs_inode_t	*ip)
+	xfs_inode_t	*ip,
+	int		flags)
 {
 	xfs_trans_t	*tp;
 	int		error;
@@ -1216,6 +1218,7 @@ xfs_inactive_free_eofblocks(
 	xfs_filblks_t	map_len;
 	int		nimaps;
 	xfs_bmbt_irec_t	imap;
+	int		use_iolock = (flags & XFS_FREE_EOF_LOCK);
 
 	/*
 	 * Figure out if there are any blocks beyond the end
@@ -1256,11 +1259,13 @@ xfs_inactive_free_eofblocks(
 		 * cache and we can't
 		 * do that within a transaction.
 		 */
-		xfs_ilock(ip, XFS_IOLOCK_EXCL);
+		if (use_iolock)
+			xfs_ilock(ip, XFS_IOLOCK_EXCL);
 		error = xfs_itruncate_start(ip, XFS_ITRUNC_DEFINITE,
 				    ip->i_size);
 		if (error) {
-			xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+			if (use_iolock)
+				xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 			return error;
 		}
 
@@ -1297,7 +1302,8 @@ xfs_inactive_free_eofblocks(
 			error = xfs_trans_commit(tp,
 						XFS_TRANS_RELEASE_LOG_RES);
 		}
-		xfs_iunlock(ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
+		xfs_iunlock(ip, (use_iolock ? (XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL)
+					    : XFS_ILOCK_EXCL));
 	}
 	return error;
 }
@@ -1573,7 +1579,8 @@ xfs_release(
 		     (ip->i_df.if_flags & XFS_IFEXTENTS))  &&
 		    (!(ip->i_d.di_flags &
 				(XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)))) {
-			if ((error = xfs_inactive_free_eofblocks(mp, ip)))
+			error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK);
+			if (error)
 				return error;
 			/* Update linux inode block count after free above */
 			vn_to_inode(vp)->i_blocks = XFS_FSB_TO_BB(mp,
@@ -1654,7 +1661,8 @@ xfs_inactive(
 		     (!(ip->i_d.di_flags &
 				(XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)) ||
 		      (ip->i_delayed_blks != 0)))) {
-			if ((error = xfs_inactive_free_eofblocks(mp, ip)))
+			error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK);
+			if (error)
 				return VN_INACTIVE_CACHE;
 			/* Update linux inode block count after free above */
 			vn_to_inode(vp)->i_blocks = XFS_FSB_TO_BB(mp,