10 files changed, 83 insertions, 110 deletions

diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index f4213ba1ff8..7f782af286b 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -131,19 +131,34 @@ xfs_file_fsync(
 {
         struct inode            *inode = file->f_mapping->host;
         struct xfs_inode        *ip = XFS_I(inode);
+        struct xfs_mount        *mp = ip->i_mount;
         struct xfs_trans        *tp;
         int                     error = 0;
         int                     log_flushed = 0;
 
         trace_xfs_file_fsync(ip);
 
-        if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+        if (XFS_FORCED_SHUTDOWN(mp))
                 return -XFS_ERROR(EIO);
 
         xfs_iflags_clear(ip, XFS_ITRUNCATED);
 
         xfs_ioend_wait(ip);
 
+        if (mp->m_flags & XFS_MOUNT_BARRIER) {
+                /*
+                 * If we have an RT and/or log subvolume we need to make sure
+                 * to flush the write cache the device used for file data
+                 * first.  This is to ensure newly written file data make
+                 * it to disk before logging the new inode size in case of
+                 * an extending write.
+                 */
+                if (XFS_IS_REALTIME_INODE(ip))
+                        xfs_blkdev_issue_flush(mp->m_rtdev_targp);
+                else if (mp->m_logdev_targp != mp->m_ddev_targp)
+                        xfs_blkdev_issue_flush(mp->m_ddev_targp);
+        }
+
         /*
          * We always need to make sure that the required inode state is safe on
          * disk.  The inode might be clean but we still might need to force the
@@ -175,9 +190,9 @@ xfs_file_fsync(
                  * updates.  The sync transaction will also force the log.
                  */
                 xfs_iunlock(ip, XFS_ILOCK_SHARED);
-                tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
+                tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
                 error = xfs_trans_reserve(tp, 0,
-                                XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
+                                XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
                 if (error) {
                         xfs_trans_cancel(tp, 0);
                         return -error;
@@ -209,28 +224,25 @@ xfs_file_fsync(
                  * force the log.
                  */
                 if (xfs_ipincount(ip)) {
-                        error = _xfs_log_force_lsn(ip->i_mount,
+                        error = _xfs_log_force_lsn(mp,
                                         ip->i_itemp->ili_last_lsn,
                                         XFS_LOG_SYNC, &log_flushed);
                 }
                 xfs_iunlock(ip, XFS_ILOCK_SHARED);
         }
 
-        if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) {
-                /*
-                 * If the log write didn't issue an ordered tag we need
-                 * to flush the disk cache for the data device now.
-                 */
-                if (!log_flushed)
-                        xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
-
-                /*
-                 * If this inode is on the RT dev we need to flush that
-                 * cache as well.
-                 */
-                if (XFS_IS_REALTIME_INODE(ip))
-                        xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
-        }
+        /*
+         * If we only have a single device, and the log force about was
+         * a no-op we might have to flush the data device cache here.
+         * This can only happen for fdatasync/O_DSYNC if we were overwriting
+         * an already allocated file and thus do not have any metadata to
+         * commit.
+         */
+        if ((mp->m_flags & XFS_MOUNT_BARRIER) &&
+            mp->m_logdev_targp == mp->m_ddev_targp &&
+            !XFS_IS_REALTIME_INODE(ip) &&
+            !log_flushed)
+                xfs_blkdev_issue_flush(mp->m_ddev_targp);
 
         return -error;
 }
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index dd21784525a..d44d92cd12b 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -182,7 +182,7 @@ xfs_vn_mknod(
         if (IS_POSIXACL(dir)) {
                 default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT);
                 if (IS_ERR(default_acl))
-                        return -PTR_ERR(default_acl);
+                        return PTR_ERR(default_acl);
 
                 if (!default_acl)
                         mode &= ~current_umask();
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 1e3a7ce804d..a1a881e68a9 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -627,68 +627,6 @@ xfs_blkdev_put(
                 blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
 }
 
-/*
- * Try to write out the superblock using barriers.
- */
-STATIC int
-xfs_barrier_test(
-        xfs_mount_t     *mp)
-{
-        xfs_buf_t       *sbp = xfs_getsb(mp, 0);
-        int             error;
-
-        XFS_BUF_UNDONE(sbp);
-        XFS_BUF_UNREAD(sbp);
-        XFS_BUF_UNDELAYWRITE(sbp);
-        XFS_BUF_WRITE(sbp);
-        XFS_BUF_UNASYNC(sbp);
-        XFS_BUF_ORDERED(sbp);
-
-        xfsbdstrat(mp, sbp);
-        error = xfs_buf_iowait(sbp);
-
-        /*
-         * Clear all the flags we set and possible error state in the
-         * buffer.  We only did the write to try out whether barriers
-         * worked and shouldn't leave any traces in the superblock
-         * buffer.
-         */
-        XFS_BUF_DONE(sbp);
-        XFS_BUF_ERROR(sbp, 0);
-        XFS_BUF_UNORDERED(sbp);
-
-        xfs_buf_relse(sbp);
-        return error;
-}
-
-STATIC void
-xfs_mountfs_check_barriers(xfs_mount_t *mp)
-{
-        int error;
-
-        if (mp->m_logdev_targp != mp->m_ddev_targp) {
-                xfs_notice(mp,
-                  "Disabling barriers, not supported with external log device");
-                mp->m_flags &= ~XFS_MOUNT_BARRIER;
-                return;
-        }
-
-        if (xfs_readonly_buftarg(mp->m_ddev_targp)) {
-                xfs_notice(mp,
-                        "Disabling barriers, underlying device is readonly");
-                mp->m_flags &= ~XFS_MOUNT_BARRIER;
-                return;
-        }
-
-        error = xfs_barrier_test(mp);
-        if (error) {
-                xfs_notice(mp,
-                        "Disabling barriers, trial barrier write failed");
-                mp->m_flags &= ~XFS_MOUNT_BARRIER;
-                return;
-        }
-}
-
 void
 xfs_blkdev_issue_flush(
         xfs_buftarg_t           *buftarg)
@@ -1240,14 +1178,6 @@ xfs_fs_remount(
                 switch (token) {
                 case Opt_barrier:
                         mp->m_flags |= XFS_MOUNT_BARRIER;
-
-                        /*
-                         * Test if barriers are actually working if we can,
-                         * else delay this check until the filesystem is
-                         * marked writeable.
-                         */
-                        if (!(mp->m_flags & XFS_MOUNT_RDONLY))
-                                xfs_mountfs_check_barriers(mp);
                         break;
                 case Opt_nobarrier:
                         mp->m_flags &= ~XFS_MOUNT_BARRIER;
@@ -1282,8 +1212,6 @@ xfs_fs_remount(
         /* ro -> rw */
         if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) {
                 mp->m_flags &= ~XFS_MOUNT_RDONLY;
-                if (mp->m_flags & XFS_MOUNT_BARRIER)
-                        xfs_mountfs_check_barriers(mp);
 
                 /*
                  * If this is the first remount to writeable state we
@@ -1465,9 +1393,6 @@ xfs_fs_fill_super(
         if (error)
                 goto out_free_sb;
 
-        if (mp->m_flags & XFS_MOUNT_BARRIER)
-                xfs_mountfs_check_barriers(mp);
-
         error = xfs_filestream_mount(mp);
         if (error)
                 goto out_free_sb;
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index c8637537881..01d2072fb6d 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -490,6 +490,13 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
         args.whichfork = XFS_ATTR_FORK;
 
         /*
+         * we have no control over the attribute names that userspace passes us
+         * to remove, so we have to allow the name lookup prior to attribute
+         * removal to fail.
+         */
+        args.op_flags = XFS_DA_OP_OKNOENT;
+
+        /*
          * Attach the dquots to the inode.
          */
         error = xfs_qm_dqattach(dp, 0);
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index cb9b6d1469f..3631783b2b5 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -253,16 +253,21 @@ xfs_iget_cache_hit(
                         rcu_read_lock();
                         spin_lock(&ip->i_flags_lock);
 
-                        ip->i_flags &= ~XFS_INEW;
-                        ip->i_flags |= XFS_IRECLAIMABLE;
-                        __xfs_inode_set_reclaim_tag(pag, ip);
+                        ip->i_flags &= ~(XFS_INEW | XFS_IRECLAIM);
+                        ASSERT(ip->i_flags & XFS_IRECLAIMABLE);
                         trace_xfs_iget_reclaim_fail(ip);
                         goto out_error;
                 }
 
                 spin_lock(&pag->pag_ici_lock);
                 spin_lock(&ip->i_flags_lock);
-                ip->i_flags &= ~(XFS_IRECLAIMABLE | XFS_IRECLAIM);
+
+                /*
+                 * Clear the per-lifetime state in the inode as we are now
+                 * effectively a new inode and need to return to the initial
+                 * state before reuse occurs.
+                 */
+                ip->i_flags &= ~XFS_IRECLAIM_RESET_FLAGS;
                 ip->i_flags |= XFS_INEW;
                 __xfs_inode_clear_reclaim_tag(mp, pag, ip);
                 inode->i_state = I_NEW;
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 3ae6d58e547..964cfea7768 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -384,6 +384,16 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
 #define XFS_IDIRTY_RELEASE      0x0040  /* dirty release already seen */
 
 /*
+ * Per-lifetime flags need to be reset when re-using a reclaimable inode during
+ * inode lookup. Thi prevents unintended behaviour on the new inode from
+ * ocurring.
+ */
+#define XFS_IRECLAIM_RESET_FLAGS        \
+        (XFS_IRECLAIMABLE | XFS_IRECLAIM | \
+         XFS_IDIRTY_RELEASE | XFS_ITRUNCATED | \
+         XFS_IFILESTREAM);
+
+/*
  * Flags for inode locking.
  * Bit ranges:  1<<1  - 1<<16-1 -- iolock/ilock modes (bitfield)
  *              1<<16 - 1<<32-1 -- lockdep annotation (integers)
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 09983a3344a..b1e88d56069 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -681,15 +681,15 @@ xfs_inode_item_unlock(
  * where the cluster buffer may be unpinned before the inode is inserted into
  * the AIL during transaction committed processing. If the buffer is unpinned
  * before the inode item has been committed and inserted, then it is possible
- * for the buffer to be written and IO completions before the inode is inserted
+ * for the buffer to be written and IO completes before the inode is inserted
  * into the AIL. In that case, we'd be inserting a clean, stale inode into the
  * AIL which will never get removed. It will, however, get reclaimed which
  * triggers an assert in xfs_inode_free() complaining about freein an inode
  * still in the AIL.
  *
- * To avoid this, return a lower LSN than the one passed in so that the
- * transaction committed code will not move the inode forward in the AIL but
- * will still unpin it properly.
+ * To avoid this, just unpin the inode directly and return a LSN of -1 so the
+ * transaction committed code knows that it does not need to do any further
+ * processing on the item.
  */
 STATIC xfs_lsn_t
 xfs_inode_item_committed(
@@ -699,8 +699,10 @@ xfs_inode_item_committed(
         struct xfs_inode_log_item *iip = INODE_ITEM(lip);
         struct xfs_inode        *ip = iip->ili_inode;
 
-        if (xfs_iflags_test(ip, XFS_ISTALE))
-                return lsn - 1;
+        if (xfs_iflags_test(ip, XFS_ISTALE)) {
+                xfs_inode_item_unpin(lip, 0);
+                return -1;
+        }
         return lsn;
 }
 
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 211930246f2..41d5b8f2bf9 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -1372,8 +1372,17 @@ xlog_sync(xlog_t		*log,
         XFS_BUF_ASYNC(bp);
         bp->b_flags |= XBF_LOG_BUFFER;
 
-        if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
+        if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) {
+                /*
+                 * If we have an external log device, flush the data device
+                 * before flushing the log to make sure all meta data
+                 * written back from the AIL actually made it to disk
+                 * before writing out the new log tail LSN in the log buffer.
+                 */
+                if (log->l_mp->m_logdev_targp != log->l_mp->m_ddev_targp)
+                        xfs_blkdev_issue_flush(log->l_mp->m_ddev_targp);
                 XFS_BUF_ORDERED(bp);
+        }
 
         ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
         ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize);
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 7c7bc2b786b..c83f63b33aa 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -1361,7 +1361,7 @@ xfs_trans_item_committed(
                 lip->li_flags |= XFS_LI_ABORTED;
         item_lsn = IOP_COMMITTED(lip, commit_lsn);
 
-        /* If the committed routine returns -1, item has been freed. */
+        /* item_lsn of -1 means the item needs no further processing */
         if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0)
                 return;
 
@@ -1474,7 +1474,7 @@ xfs_trans_committed_bulk(
                         lip->li_flags |= XFS_LI_ABORTED;
                 item_lsn = IOP_COMMITTED(lip, commit_lsn);
 
-                /* item_lsn of -1 means the item was freed */
+                /* item_lsn of -1 means the item needs no further processing */
                 if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0)
                         continue;
 
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index b7a5fe7c52c..619720705bc 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -960,8 +960,11 @@ xfs_release(
                  * be exposed to that problem.
                  */
                 truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED);
-                if (truncated && VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0)
-                        xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE);
+                if (truncated) {
+                        xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE);
+                        if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0)
+                                xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE);
+                }
         }
 
         if (ip->i_d.di_nlink == 0)