Merge remote branch 'xfs/master'

14 files changed, 135 insertions, 173 deletions

diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index c2e30eea74dc..616ee3febcbb 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -235,71 +235,36 @@ xfs_setfilesize(
 }
 
 /*
- * Buffered IO write completion for delayed allocate extents.
+ * IO write completion.
  */
 STATIC void
-xfs_end_bio_delalloc(
-	struct work_struct	*work)
-{
-	xfs_ioend_t		*ioend =
-		container_of(work, xfs_ioend_t, io_work);
-
-	xfs_setfilesize(ioend);
-	xfs_destroy_ioend(ioend);
-}
-
-/*
- * Buffered IO write completion for regular, written extents.
- */
-STATIC void
-xfs_end_bio_written(
-	struct work_struct	*work)
-{
-	xfs_ioend_t		*ioend =
-		container_of(work, xfs_ioend_t, io_work);
-
-	xfs_setfilesize(ioend);
-	xfs_destroy_ioend(ioend);
-}
-
-/*
- * IO write completion for unwritten extents.
- *
- * Issue transactions to convert a buffer range from unwritten
- * to written extents.
- */
-STATIC void
-xfs_end_bio_unwritten(
+xfs_end_io(
 	struct work_struct	*work)
 {
 	xfs_ioend_t		*ioend =
 		container_of(work, xfs_ioend_t, io_work);
 	struct xfs_inode	*ip = XFS_I(ioend->io_inode);
-	xfs_off_t		offset = ioend->io_offset;
-	size_t			size = ioend->io_size;
-
-	if (likely(!ioend->io_error)) {
-		if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-			int error;
-			error = xfs_iomap_write_unwritten(ip, offset, size);
-			if (error)
-				ioend->io_error = error;
-		}
-		xfs_setfilesize(ioend);
-	}
-	xfs_destroy_ioend(ioend);
-}
 
-/*
- * IO read completion for regular, written extents.
- */
-STATIC void
-xfs_end_bio_read(
-	struct work_struct	*work)
-{
-	xfs_ioend_t		*ioend =
-		container_of(work, xfs_ioend_t, io_work);
+	/*
+	 * For unwritten extents we need to issue transactions to convert a
+	 * range to normal written extens after the data I/O has finished.
+	 */
+	if (ioend->io_type == IOMAP_UNWRITTEN &&
+	    likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) {
+		int error;
+
+		error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
+						 ioend->io_size);
+		if (error)
+			ioend->io_error = error;
+	}
 
+	/*
+	 * We might have to update the on-disk file size after extending
+	 * writes.
+	 */
+	if (ioend->io_type != IOMAP_READ)
+		xfs_setfilesize(ioend);
 	xfs_destroy_ioend(ioend);
 }
 
@@ -314,10 +279,10 @@ xfs_finish_ioend(
 	int		wait)
 {
 	if (atomic_dec_and_test(&ioend->io_remaining)) {
-		struct workqueue_struct *wq = xfsdatad_workqueue;
-		if (ioend->io_work.func == xfs_end_bio_unwritten)
-			wq = xfsconvertd_workqueue;
+		struct workqueue_struct *wq;
 
+		wq = (ioend->io_type == IOMAP_UNWRITTEN) ?
+			xfsconvertd_workqueue : xfsdatad_workqueue;
 		queue_work(wq, &ioend->io_work);
 		if (wait)
 			flush_workqueue(wq);
@@ -355,15 +320,7 @@ xfs_alloc_ioend(
 	ioend->io_offset = 0;
 	ioend->io_size = 0;
 
-	if (type == IOMAP_UNWRITTEN)
-		INIT_WORK(&ioend->io_work, xfs_end_bio_unwritten);
-	else if (type == IOMAP_DELAY)
-		INIT_WORK(&ioend->io_work, xfs_end_bio_delalloc);
-	else if (type == IOMAP_READ)
-		INIT_WORK(&ioend->io_work, xfs_end_bio_read);
-	else
-		INIT_WORK(&ioend->io_work, xfs_end_bio_written);
-
+	INIT_WORK(&ioend->io_work, xfs_end_io);
 	return ioend;
 }
 
@@ -412,8 +369,9 @@ xfs_end_bio(
 
 STATIC void
 xfs_submit_ioend_bio(
-	xfs_ioend_t	*ioend,
-	struct bio	*bio)
+	struct writeback_control *wbc,
+	xfs_ioend_t		*ioend,
+	struct bio		*bio)
 {
 	atomic_inc(&ioend->io_remaining);
 	bio->bi_private = ioend;
@@ -426,7 +384,8 @@ xfs_submit_ioend_bio(
 	if (xfs_ioend_new_eof(ioend))
 		xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode));
 
-	submit_bio(WRITE, bio);
+	submit_bio(wbc->sync_mode == WB_SYNC_ALL ?
+		   WRITE_SYNC_PLUG : WRITE, bio);
 	ASSERT(!bio_flagged(bio, BIO_EOPNOTSUPP));
 	bio_put(bio);
 }
@@ -505,6 +464,7 @@ static inline int bio_add_buffer(struct bio *bio, struct buffer_head *bh)
  */
 STATIC void
 xfs_submit_ioend(
+	struct writeback_control *wbc,
 	xfs_ioend_t		*ioend)
 {
 	xfs_ioend_t		*head = ioend;
@@ -533,19 +493,19 @@ xfs_submit_ioend(
  retry:
 				bio = xfs_alloc_ioend_bio(bh);
 			} else if (bh->b_blocknr != lastblock + 1) {
-				xfs_submit_ioend_bio(ioend, bio);
+				xfs_submit_ioend_bio(wbc, ioend, bio);
 				goto retry;
 			}
 
 			if (bio_add_buffer(bio, bh) != bh->b_size) {
-				xfs_submit_ioend_bio(ioend, bio);
+				xfs_submit_ioend_bio(wbc, ioend, bio);
 				goto retry;
 			}
 
 			lastblock = bh->b_blocknr;
 		}
 		if (bio)
-			xfs_submit_ioend_bio(ioend, bio);
+			xfs_submit_ioend_bio(wbc, ioend, bio);
 		xfs_finish_ioend(ioend, 0);
 	} while ((ioend = next) != NULL);
 }
@@ -1198,7 +1158,7 @@ xfs_page_state_convert(
 	}
 
 	if (iohead)
-		xfs_submit_ioend(iohead);
+		xfs_submit_ioend(wbc, iohead);
 
 	return page_dirty;
 
@@ -1535,7 +1495,7 @@ xfs_end_io_direct(
 		 * didn't map an unwritten extent so switch it's completion
 		 * handler.
 		 */
-		INIT_WORK(&ioend->io_work, xfs_end_bio_written);
+		ioend->io_type = IOMAP_NEW;
 		xfs_finish_ioend(ioend, 0);
 	}
 
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 18a4b8e11df2..25b6903a3bce 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -930,13 +930,39 @@ xfs_fs_alloc_inode(
  */
 STATIC void
 xfs_fs_destroy_inode(
-	struct inode	*inode)
+	struct inode		*inode)
 {
-	xfs_inode_t		*ip = XFS_I(inode);
+	struct xfs_inode	*ip = XFS_I(inode);
+
+	xfs_itrace_entry(ip);
 
 	XFS_STATS_INC(vn_reclaim);
-	if (xfs_reclaim(ip))
-		panic("%s: cannot reclaim 0x%p\n", __func__, inode);
+
+	/* bad inode, get out here ASAP */
+	if (is_bad_inode(inode))
+		goto out_reclaim;
+
+	xfs_ioend_wait(ip);
+
+	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
+
+	/*
+	 * We should never get here with one of the reclaim flags already set.
+	 */
+	ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIMABLE));
+	ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM));
+
+	/*
+	 * If we have nothing to flush with this inode then complete the
+	 * teardown now, otherwise delay the flush operation.
+	 */
+	if (!xfs_inode_clean(ip)) {
+		xfs_inode_set_reclaim_tag(ip);
+		return;
+	}
+
+out_reclaim:
+	xfs_ireclaim(ip);
 }
 
 /*
@@ -973,7 +999,6 @@ xfs_fs_inode_init_once(
 
 	mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER,
 		     "xfsino", ip->i_ino);
-	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
 }
 
 /*
@@ -1075,6 +1100,20 @@ xfs_fs_clear_inode(
 	XFS_STATS_INC(vn_remove);
 	XFS_STATS_DEC(vn_active);
 
+	/*
+	 * The iolock is used by the file system to coordinate reads,
+	 * writes, and block truncates.  Up to this point the lock
+	 * protected concurrent accesses by users of the inode.  But
+	 * from here forward we're doing some final processing of the
+	 * inode because we're done with it, and although we reuse the
+	 * iolock for protection it is really a distinct lock class
+	 * (in the lockdep sense) from before.  To keep lockdep happy
+	 * (and basically indicate what we are doing), we explicitly
+	 * re-init the iolock here.
+	 */
+	ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock));
+	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
+
 	xfs_inactive(ip);
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 961df0a22c78..d895a3a960f5 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -663,10 +663,9 @@ xfs_syncd_stop(
 	kthread_stop(mp->m_sync_task);
 }
 
-int
+STATIC int
 xfs_reclaim_inode(
 	xfs_inode_t	*ip,
-	int		locked,
 	int		sync_mode)
 {
 	xfs_perag_t	*pag = xfs_get_perag(ip->i_mount, ip->i_ino);
@@ -682,10 +681,6 @@ xfs_reclaim_inode(
 	    !__xfs_iflags_test(ip, XFS_IRECLAIMABLE)) {
 		spin_unlock(&ip->i_flags_lock);
 		write_unlock(&pag->pag_ici_lock);
-		if (locked) {
-			xfs_ifunlock(ip);
-			xfs_iunlock(ip, XFS_ILOCK_EXCL);
-		}
 		return -EAGAIN;
 	}
 	__xfs_iflags_set(ip, XFS_IRECLAIM);
@@ -704,10 +699,8 @@ xfs_reclaim_inode(
 	 * We get the flush lock regardless, though, just to make sure
 	 * we don't free it while it is being flushed.
 	 */
-	if (!locked) {
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_iflock(ip);
-	}
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_iflock(ip);
 
 	/*
 	 * In the case of a forced shutdown we rely on xfs_iflush() to
@@ -778,7 +771,7 @@ xfs_reclaim_inode_now(
 	}
 	read_unlock(&pag->pag_ici_lock);
 
-	return xfs_reclaim_inode(ip, 0, flags);
+	return xfs_reclaim_inode(ip, flags);
 }
 
 int
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index 27920eb7a820..a500b4d91835 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -44,7 +44,6 @@ void xfs_quiesce_attr(struct xfs_mount *mp);
 
 void xfs_flush_inodes(struct xfs_inode *ip);
 
-int xfs_reclaim_inode(struct xfs_inode *ip, int locked, int sync_mode);
 int xfs_reclaim_inodes(struct xfs_mount *mp, int mode);
 
 void xfs_inode_set_reclaim_tag(struct xfs_inode *ip);
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 2d0b3e1da9e6..6f83f58c099f 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -611,7 +611,7 @@ xfs_fs_log_dummy(
 	xfs_inode_t	*ip;
 	int		error;
 
-	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1);
+	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1, KM_SLEEP);
 	error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 80e526489be5..cc72c561ff52 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -73,6 +73,9 @@ xfs_inode_alloc(
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
 	ASSERT(completion_done(&ip->i_flush));
+	ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock));
+
+	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
 
 	/* initialise the xfs inode */
 	ip->i_ino = ino;
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 67ae5555a30a..7294abce6ef2 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -860,8 +860,15 @@ xfs_iomap_write_unwritten(
 		 * set up a transaction to convert the range of extents
 		 * from unwritten to real. Do allocations in a loop until
 		 * we have covered the range passed in.
+		 *
+		 * Note that we open code the transaction allocation here
+		 * to pass KM_NOFS--we can't risk to recursing back into
+		 * the filesystem here as we might be asked to write out
+		 * the same inode that we complete here and might deadlock
+		 * on the iolock.
 		 */
-		tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
+		xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
+		tp = _xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE, KM_NOFS);
 		tp->t_flags |= XFS_TRANS_RESERVE;
 		error = xfs_trans_reserve(tp, resblks,
 				XFS_WRITE_LOG_RES(mp), 0,
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index fb17f8226b09..b5b0d8055910 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -3517,7 +3517,7 @@ xlog_do_recovery_pass(
 {
 	xlog_rec_header_t	*rhead;
 	xfs_daddr_t		blk_no;
-	xfs_caddr_t		bufaddr, offset;
+	xfs_caddr_t		offset;
 	xfs_buf_t		*hbp, *dbp;
 	int			error = 0, h_size;
 	int			bblks, split_bblks;
@@ -3610,7 +3610,7 @@ xlog_do_recovery_pass(
 			/*
 			 * Check for header wrapping around physical end-of-log
 			 */
-			offset = NULL;
+			offset = XFS_BUF_PTR(hbp);
 			split_hblks = 0;
 			wrapped_hblks = 0;
 			if (blk_no + hblks <= log->l_logBBsize) {
@@ -3646,9 +3646,8 @@ xlog_do_recovery_pass(
 				 *   - order is important.
 				 */
 				wrapped_hblks = hblks - split_hblks;
-				bufaddr = XFS_BUF_PTR(hbp);
 				error = XFS_BUF_SET_PTR(hbp,
-						bufaddr + BBTOB(split_hblks),
+						offset + BBTOB(split_hblks),
 						BBTOB(hblks - split_hblks));
 				if (error)
 					goto bread_err2;
@@ -3658,14 +3657,10 @@ xlog_do_recovery_pass(
 				if (error)
 					goto bread_err2;
 
-				error = XFS_BUF_SET_PTR(hbp, bufaddr,
+				error = XFS_BUF_SET_PTR(hbp, offset,
 							BBTOB(hblks));
 				if (error)
 					goto bread_err2;
-
-				if (!offset)
-					offset = xlog_align(log, 0,
-							wrapped_hblks, hbp);
 			}
 			rhead = (xlog_rec_header_t *)offset;
 			error = xlog_valid_rec_header(log, rhead,
@@ -3685,7 +3680,7 @@ xlog_do_recovery_pass(
 			} else {
 				/* This log record is split across the
 				 * physical end of log */
-				offset = NULL;
+				offset = XFS_BUF_PTR(dbp);
 				split_bblks = 0;
 				if (blk_no != log->l_logBBsize) {
 					/* some data is before the physical
@@ -3714,9 +3709,8 @@ xlog_do_recovery_pass(
 				 *   _first_, then the log start (LR header end)
 				 *   - order is important.
 				 */
-				bufaddr = XFS_BUF_PTR(dbp);
 				error = XFS_BUF_SET_PTR(dbp,
-						bufaddr + BBTOB(split_bblks),
+						offset + BBTOB(split_bblks),
 						BBTOB(bblks - split_bblks));
 				if (error)
 					goto bread_err2;
@@ -3727,13 +3721,9 @@ xlog_do_recovery_pass(
 				if (error)
 					goto bread_err2;
 
-				error = XFS_BUF_SET_PTR(dbp, bufaddr, h_size);
+				error = XFS_BUF_SET_PTR(dbp, offset, h_size);
 				if (error)
 					goto bread_err2;
-
-				if (!offset)
-					offset = xlog_align(log, wrapped_hblks,
-						bblks - split_bblks, dbp);
 			}
 			xlog_unpack_data(rhead, offset, log);
 			if ((error = xlog_recover_process_data(log, rhash,
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 8b6c9e807efb..4d509f742bd2 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -1471,7 +1471,7 @@ xfs_log_sbcount(
 	if (!xfs_sb_version_haslazysbcount(&mp->m_sb))
 		return 0;
 
-	tp = _xfs_trans_alloc(mp, XFS_TRANS_SB_COUNT);
+	tp = _xfs_trans_alloc(mp, XFS_TRANS_SB_COUNT, KM_SLEEP);
 	error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
 					XFS_DEFAULT_LOG_COUNT);
 	if (error) {
diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h
index f5e4874c37d8..726014d1c925 100644
--- a/fs/xfs/xfs_rw.h
+++ b/fs/xfs/xfs_rw.h
@@ -37,13 +37,6 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
 }
 
 /*
- * Flags for xfs_free_eofblocks
- */
-#define XFS_FREE_EOF_LOCK	(1<<0)
-#define XFS_FREE_EOF_NOLOCK	(1<<1)
-
-
-/*
  * helper function to extract extent size hint from inode
  */
 STATIC_INLINE xfs_extlen_t
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 66b849358e62..237badcbac3b 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -236,19 +236,20 @@ xfs_trans_alloc(
 	uint		type)
 {
 	xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
-	return _xfs_trans_alloc(mp, type);
+	return _xfs_trans_alloc(mp, type, KM_SLEEP);
 }
 
 xfs_trans_t *
 _xfs_trans_alloc(
 	xfs_mount_t	*mp,
-	uint		type)
+	uint		type,
+	uint		memflags)
 {
 	xfs_trans_t	*tp;
 
 	atomic_inc(&mp->m_active_trans);
 
-	tp = kmem_zone_zalloc(xfs_trans_zone, KM_SLEEP);
+	tp = kmem_zone_zalloc(xfs_trans_zone, memflags);
 	tp->t_magic = XFS_TRANS_MAGIC;
 	tp->t_type = type;
 	tp->t_mountp = mp;
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index ed47fc77759c..a0574f593f52 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -924,7 +924,7 @@ typedef struct xfs_trans {
  * XFS transaction mechanism exported interfaces.
  */
 xfs_trans_t	*xfs_trans_alloc(struct xfs_mount *, uint);
-xfs_trans_t	*_xfs_trans_alloc(struct xfs_mount *, uint);
+xfs_trans_t	*_xfs_trans_alloc(struct xfs_mount *, uint, uint);
 xfs_trans_t	*xfs_trans_dup(xfs_trans_t *);
 int		xfs_trans_reserve(xfs_trans_t *, uint, uint, uint,
 				  uint, uint);
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index b572f7e840e0..d98401470cf0 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -709,6 +709,11 @@ xfs_fsync(
 }
 
 /*
+ * Flags for xfs_free_eofblocks
+ */
+#define XFS_FREE_EOF_TRYLOCK	(1<<0)
+
+/*
  * 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.
@@ -726,7 +731,6 @@ xfs_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
@@ -768,14 +772,19 @@ xfs_free_eofblocks(
 		 * cache and we can't
 		 * do that within a transaction.
 		 */
-		if (use_iolock)
+		if (flags & XFS_FREE_EOF_TRYLOCK) {
+			if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) {
+				xfs_trans_cancel(tp, 0);
+				return 0;
+			}
+		} else {
 			xfs_ilock(ip, XFS_IOLOCK_EXCL);
+		}
 		error = xfs_itruncate_start(ip, XFS_ITRUNC_DEFINITE,
 				    ip->i_size);
 		if (error) {
 			xfs_trans_cancel(tp, 0);
-			if (use_iolock)
-				xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+			xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 			return error;
 		}
 
@@ -812,8 +821,7 @@ xfs_free_eofblocks(
 			error = xfs_trans_commit(tp,
 						XFS_TRANS_RELEASE_LOG_RES);
 		}
-		xfs_iunlock(ip, (use_iolock ? (XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL)
-					    : XFS_ILOCK_EXCL));
+		xfs_iunlock(ip, XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL);
 	}
 	return error;
 }
@@ -1113,7 +1121,17 @@ xfs_release(
 		     (ip->i_df.if_flags & XFS_IFEXTENTS))  &&
 		    (!(ip->i_d.di_flags &
 				(XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)))) {
-			error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK);
+
+			/*
+			 * If we can't get the iolock just skip truncating
+			 * the blocks past EOF because we could deadlock
+			 * with the mmap_sem otherwise.  We'll get another
+			 * chance to drop them once the last reference to
+			 * the inode is dropped, so we'll never leak blocks
+			 * permanently.
+			 */
+			error = xfs_free_eofblocks(mp, ip,
+						   XFS_FREE_EOF_TRYLOCK);
 			if (error)
 				return error;
 		}
@@ -1184,7 +1202,7 @@ xfs_inactive(
 		     (!(ip->i_d.di_flags &
 				(XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)) ||
 		      (ip->i_delayed_blks != 0)))) {
-			error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK);
+			error = xfs_free_eofblocks(mp, ip, 0);
 			if (error)
 				return VN_INACTIVE_CACHE;
 		}
@@ -2456,46 +2474,6 @@ xfs_set_dmattrs(
 	return error;
 }
 
-int
-xfs_reclaim(
-	xfs_inode_t	*ip)
-{
-
-	xfs_itrace_entry(ip);
-
-	ASSERT(!VN_MAPPED(VFS_I(ip)));
-
-	/* bad inode, get out here ASAP */
-	if (is_bad_inode(VFS_I(ip))) {
-		xfs_ireclaim(ip);
-		return 0;
-	}
-
-	xfs_ioend_wait(ip);
-
-	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
-
-	/*
-	 * If we have nothing to flush with this inode then complete the
-	 * teardown now, otherwise break the link between the xfs inode and the
-	 * linux inode and clean up the xfs inode later. This avoids flushing
-	 * the inode to disk during the delete operation itself.
-	 *
-	 * When breaking the link, we need to set the XFS_IRECLAIMABLE flag
-	 * first to ensure that xfs_iunpin() will never see an xfs inode
-	 * that has a linux inode being reclaimed. Synchronisation is provided
-	 * by the i_flags_lock.
-	 */
-	if (!ip->i_update_core && (ip->i_itemp == NULL)) {
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_iflock(ip);
-		xfs_iflags_set(ip, XFS_IRECLAIMABLE);
-		return xfs_reclaim_inode(ip, 1, XFS_IFLUSH_DELWRI_ELSE_SYNC);
-	}
-	xfs_inode_set_reclaim_tag(ip);
-	return 0;
-}
-
 /*
  * xfs_alloc_file_space()
  *      This routine allocates disk space for the given file.
diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h
index a9e102de71a1..167a467403a5 100644
--- a/fs/xfs/xfs_vnodeops.h
+++ b/fs/xfs/xfs_vnodeops.h
@@ -38,7 +38,6 @@ int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
 		const char *target_path, mode_t mode, struct xfs_inode **ipp,
 		cred_t *credp);
 int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state);
-int xfs_reclaim(struct xfs_inode *ip);
 int xfs_change_file_space(struct xfs_inode *ip, int cmd,
 		xfs_flock64_t *bf, xfs_off_t offset, int attr_flags);
 int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name,