Merge branch 'for-linus' of git://oss.sgi.com/xfs/xfs

* 'for-linus' of git://oss.sgi.com/xfs/xfs:
  xfs: Fix error return for fallocate() on XFS
  xfs: cleanup dmapi macros in the umount path
  xfs: remove incorrect sparse annotation for xfs_iget_cache_miss
  xfs: kill the STATIC_INLINE macro
  xfs: uninline xfs_get_extsz_hint
  xfs: rename xfs_attr_fetch to xfs_attr_get_int
  xfs: simplify xfs_buf_get / xfs_buf_read interfaces
  xfs: remove IO_ISAIO
  xfs: Wrapped journal record corruption on read at recovery
  xfs: cleanup data end I/O handlers
  xfs: use WRITE_SYNC_PLUG for synchronous writeout
  xfs: reset the i_iolock lock class in the reclaim path
  xfs: I/O completion handlers must use NOFS allocations
  xfs: fix mmap_sem/iolock inversion in xfs_free_eofblocks
  xfs: simplify inode teardown

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;
 }
 
@@ -380,7 +337,7 @@ xfs_map_blocks(
 	return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
 }
 
-STATIC_INLINE int
+STATIC int
 xfs_iomap_valid(
 	xfs_iomap_t		*iomapp,
 	loff_t			offset)
@@ -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);
 }
@@ -1191,7 +1151,7 @@ xfs_page_state_convert(
 	}
 
 	if (iohead)
-		xfs_submit_ioend(iohead);
+		xfs_submit_ioend(wbc, iohead);
 
 	return page_dirty;
 
@@ -1528,7 +1488,7 @@ 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);
 	}
 

fs/xfs/linux-2.6/xfs_buf.c

@@ -149,7 +149,7 @@ page_region_mask(
 	return mask;
 }
 
-STATIC_INLINE void
+STATIC void
 set_page_region(
 	struct page	*page,
 	size_t		offset,
@@ -161,7 +161,7 @@ set_page_region(
 		SetPageUptodate(page);
 }
 
-STATIC_INLINE int
+STATIC int
 test_page_region(
 	struct page	*page,
 	size_t		offset,
@@ -582,7 +582,7 @@ found:
  *	although backing storage may not be.
  */
 xfs_buf_t *
-xfs_buf_get_flags(
+xfs_buf_get(
 	xfs_buftarg_t		*target,/* target for buffer		*/
 	xfs_off_t		ioff,	/* starting offset of range	*/
 	size_t			isize,	/* length of range		*/
@@ -661,7 +661,7 @@ _xfs_buf_read(
 }
 
 xfs_buf_t *
-xfs_buf_read_flags(
+xfs_buf_read(
 	xfs_buftarg_t		*target,
 	xfs_off_t		ioff,
 	size_t			isize,
@@ -671,7 +671,7 @@ xfs_buf_read_flags(
 
 	flags |= XBF_READ;
 
-	bp = xfs_buf_get_flags(target, ioff, isize, flags);
+	bp = xfs_buf_get(target, ioff, isize, flags);
 	if (bp) {
 		if (!XFS_BUF_ISDONE(bp)) {
 			XB_TRACE(bp, "read", (unsigned long)flags);
@@ -718,7 +718,7 @@ xfs_buf_readahead(
 		return;
 
 	flags |= (XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD);
-	xfs_buf_read_flags(target, ioff, isize, flags);
+	xfs_buf_read(target, ioff, isize, flags);
 }
 
 xfs_buf_t *
@@ -1113,7 +1113,7 @@ xfs_bdwrite(
 	xfs_buf_delwri_queue(bp, 1);
 }
 
-STATIC_INLINE void
+STATIC void
 _xfs_buf_ioend(
 	xfs_buf_t		*bp,
 	int			schedule)

fs/xfs/linux-2.6/xfs_buf.h

@@ -186,15 +186,10 @@ extern xfs_buf_t *_xfs_buf_find(xfs_buftarg_t *, xfs_off_t, size_t,
 #define xfs_incore(buftarg,blkno,len,lockit) \
 	_xfs_buf_find(buftarg, blkno ,len, lockit, NULL)
 
-extern xfs_buf_t *xfs_buf_get_flags(xfs_buftarg_t *, xfs_off_t, size_t,
+extern xfs_buf_t *xfs_buf_get(xfs_buftarg_t *, xfs_off_t, size_t,
 				xfs_buf_flags_t);
-#define xfs_buf_get(target, blkno, len, flags) \
-	xfs_buf_get_flags((target), (blkno), (len), XBF_LOCK | XBF_MAPPED)
-
-extern xfs_buf_t *xfs_buf_read_flags(xfs_buftarg_t *, xfs_off_t, size_t,
+extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t,
 				xfs_buf_flags_t);
-#define xfs_buf_read(target, blkno, len, flags) \
-	xfs_buf_read_flags((target), (blkno), (len), XBF_LOCK | XBF_MAPPED)
 
 extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *);
 extern xfs_buf_t *xfs_buf_get_noaddr(size_t, xfs_buftarg_t *);

fs/xfs/linux-2.6/xfs_file.c

@@ -52,7 +52,7 @@ xfs_file_aio_read(
 	loff_t			pos)
 {
 	struct file		*file = iocb->ki_filp;
-	int			ioflags = IO_ISAIO;
+	int			ioflags = 0;
 
 	BUG_ON(iocb->ki_pos != pos);
 	if (unlikely(file->f_flags & O_DIRECT))
@@ -71,7 +71,7 @@ xfs_file_aio_write(
 	loff_t			pos)
 {
 	struct file		*file = iocb->ki_filp;
-	int			ioflags = IO_ISAIO;
+	int			ioflags = 0;
 
 	BUG_ON(iocb->ki_pos != pos);
 	if (unlikely(file->f_flags & O_DIRECT))

fs/xfs/linux-2.6/xfs_iops.c

@@ -573,8 +573,8 @@ xfs_vn_fallocate(
 	bf.l_len = len;
 
 	xfs_ilock(ip, XFS_IOLOCK_EXCL);
-	error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
-				      0, XFS_ATTR_NOLOCK);
+	error = -xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
+				       0, XFS_ATTR_NOLOCK);
 	if (!error && !(mode & FALLOC_FL_KEEP_SIZE) &&
 	    offset + len > i_size_read(inode))
 		new_size = offset + len;
@@ -585,7 +585,7 @@ xfs_vn_fallocate(
 
 		iattr.ia_valid = ATTR_SIZE;
 		iattr.ia_size = new_size;
-		error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
+		error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
 	}
 
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);

fs/xfs/linux-2.6/xfs_lrw.c

@@ -255,8 +255,6 @@ xfs_read(
 
 	iocb->ki_pos = *offset;
 	ret = generic_file_aio_read(iocb, iovp, segs, *offset);
-	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
-		ret = wait_on_sync_kiocb(iocb);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_read_bytes, ret);
 
@@ -774,9 +772,6 @@ write_retry:
 
 	current->backing_dev_info = NULL;
 
-	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
-		ret = wait_on_sync_kiocb(iocb);
-
 	isize = i_size_read(inode);
 	if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize))
 		*offset = isize;

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);
 }
 
@@ -1092,8 +1131,6 @@ xfs_fs_put_super(
 	struct super_block	*sb)
 {
 	struct xfs_mount	*mp = XFS_M(sb);
-	struct xfs_inode	*rip = mp->m_rootip;
-	int			unmount_event_flags = 0;
 
 	xfs_syncd_stop(mp);
 
@@ -1109,20 +1146,7 @@ xfs_fs_put_super(
 		xfs_sync_attr(mp, 0);
 	}
 
-#ifdef HAVE_DMAPI
-	if (mp->m_flags & XFS_MOUNT_DMAPI) {
-		unmount_event_flags =
-			(mp->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ?
-				0 : DM_FLAGS_UNWANTED;
-		/*
-		 * Ignore error from dmapi here, first unmount is not allowed
-		 * to fail anyway, and second we wouldn't want to fail a
-		 * unmount because of dmapi.
-		 */
-		XFS_SEND_PREUNMOUNT(mp, rip, DM_RIGHT_NULL, rip, DM_RIGHT_NULL,
-				NULL, NULL, 0, 0, unmount_event_flags);
-	}
-#endif
+	XFS_SEND_PREUNMOUNT(mp);
 
 	/*
 	 * Blow away any referenced inode in the filestreams cache.
@@ -1133,10 +1157,7 @@ xfs_fs_put_super(
 
 	XFS_bflush(mp->m_ddev_targp);
 
-	if (mp->m_flags & XFS_MOUNT_DMAPI) {
-		XFS_SEND_UNMOUNT(mp, rip, DM_RIGHT_NULL, 0, 0,
-				unmount_event_flags);
-	}
+	XFS_SEND_UNMOUNT(mp);
 
 	xfs_unmountfs(mp);
 	xfs_freesb(mp);

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

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);

fs/xfs/linux-2.6/xfs_vnode.h

@@ -36,7 +36,6 @@ struct attrlist_cursor_kern;
 /*
  * Flags for read/write calls - same values as IRIX
  */
-#define IO_ISAIO	0x00001		/* don't wait for completion */
 #define IO_ISDIRECT	0x00004		/* bypass page cache */
 #define IO_INVIS	0x00020		/* don't update inode timestamps */
 

fs/xfs/support/debug.h

@@ -41,10 +41,6 @@ extern void assfail(char *expr, char *f, int l);
 # define STATIC static noinline
 #endif
 
-#ifndef STATIC_INLINE
-# define STATIC_INLINE static inline
-#endif
-
 #else /* DEBUG */
 
 #define ASSERT(expr)	\
@@ -54,19 +50,5 @@ extern void assfail(char *expr, char *f, int l);
 # define STATIC noinline
 #endif
 
-/*
- * We stop inlining of inline functions in debug mode.
- * Unfortunately, this means static inline in header files
- * get multiple definitions, so they need to remain static.
- * This then gives tonnes of warnings about unused but defined
- * functions, so we need to add the unused attribute to prevent
- * these spurious warnings.
- */
-#ifndef STATIC_INLINE
-# define STATIC_INLINE static __attribute__ ((unused)) noinline
-#endif
-
 #endif /* DEBUG */
-
-
 #endif  /* __XFS_SUPPORT_DEBUG_H__ */

fs/xfs/xfs_attr.c

@@ -123,9 +123,13 @@ xfs_inode_hasattr(
  * Overall external interface routines.
  *========================================================================*/
 
-int
-xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name,
-		char *value, int *valuelenp, int flags)
+STATIC int
+xfs_attr_get_int(
+	struct xfs_inode	*ip,
+	struct xfs_name		*name,
+	char			*value,
+	int			*valuelenp,
+	int			flags)
 {
 	xfs_da_args_t   args;
 	int             error;
@@ -188,7 +192,7 @@ xfs_attr_get(
 		return error;
 
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	error = xfs_attr_fetch(ip, &xname, value, valuelenp, flags);
+	error = xfs_attr_get_int(ip, &xname, value, valuelenp, flags);
 	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 	return(error);
 }
@@ -2143,8 +2147,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
 		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
 		blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
 
-		bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, blkcnt,
-				       XFS_BUF_LOCK | XBF_DONT_BLOCK);
+		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt,
+				 XFS_BUF_LOCK | XBF_DONT_BLOCK);
 		ASSERT(bp);
 		ASSERT(!XFS_BUF_GETERROR(bp));
 

fs/xfs/xfs_attr.h

@@ -131,7 +131,6 @@ typedef struct xfs_attr_list_context {
  */
 int xfs_attr_calc_size(struct xfs_inode *, int, int, int *);
 int xfs_attr_inactive(struct xfs_inode *dp);
-int xfs_attr_fetch(struct xfs_inode *, struct xfs_name *, char *, int *, int);
 int xfs_attr_rmtval_get(struct xfs_da_args *args);
 int xfs_attr_list_int(struct xfs_attr_list_context *);
 

fs/xfs/xfs_attr_leaf.c

@@ -98,7 +98,7 @@ STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index);
  * If namespace bits don't match return 0.
  * If all match then return 1.
  */
-STATIC_INLINE int
+STATIC int
 xfs_attr_namesp_match(int arg_flags, int ondisk_flags)
 {
 	return XFS_ATTR_NSP_ONDISK(ondisk_flags) == XFS_ATTR_NSP_ARGS_TO_ONDISK(arg_flags);

fs/xfs/xfs_bmap_btree.c

@@ -98,8 +98,7 @@ xfs_bmdr_to_bmbt(
  * This code must be in sync with the routines xfs_bmbt_get_startoff,
  * xfs_bmbt_get_startblock, xfs_bmbt_get_blockcount and xfs_bmbt_get_state.
  */
-
-STATIC_INLINE void
+STATIC void
 __xfs_bmbt_get_all(
 		__uint64_t l0,
 		__uint64_t l1,

fs/xfs/xfs_filestream.h

@@ -79,7 +79,7 @@ extern ktrace_t *xfs_filestreams_trace_buf;
  * the cache that reference per-ag array elements that have since been
  * reallocated.
  */
-STATIC_INLINE int
+static inline int
 xfs_filestream_peek_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -87,7 +87,7 @@ xfs_filestream_peek_ag(
 	return atomic_read(&mp->m_perag[agno].pagf_fstrms);
 }
 
-STATIC_INLINE int
+static inline int
 xfs_filestream_get_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -95,7 +95,7 @@ xfs_filestream_get_ag(
 	return atomic_inc_return(&mp->m_perag[agno].pagf_fstrms);
 }
 
-STATIC_INLINE int
+static inline int
 xfs_filestream_put_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -122,7 +122,7 @@ int xfs_filestream_new_ag(struct xfs_bmalloca *ap, xfs_agnumber_t *agp);
 
 
 /* filestreams for the inode? */
-STATIC_INLINE int
+static inline int
 xfs_inode_is_filestream(
 	struct xfs_inode	*ip)
 {

fs/xfs/xfs_fsops.c

@@ -201,8 +201,8 @@ xfs_growfs_data_private(
 		 * AG freelist header block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-				  XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
-				  XFS_FSS_TO_BB(mp, 1), 0);
+				 XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
+				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
 		agf = XFS_BUF_TO_AGF(bp);
 		memset(agf, 0, mp->m_sb.sb_sectsize);
 		agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC);
@@ -233,8 +233,8 @@ xfs_growfs_data_private(
 		 * AG inode header block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-				  XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
-				  XFS_FSS_TO_BB(mp, 1), 0);
+				 XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
+				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
 		agi = XFS_BUF_TO_AGI(bp);
 		memset(agi, 0, mp->m_sb.sb_sectsize);
 		agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC);
@@ -257,8 +257,9 @@ xfs_growfs_data_private(
 		 * BNO btree root block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-			XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
-			BTOBB(mp->m_sb.sb_blocksize), 0);
+				 XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
+				 BTOBB(mp->m_sb.sb_blocksize),
+				 XBF_LOCK | XBF_MAPPED);
 		block = XFS_BUF_TO_BLOCK(bp);
 		memset(block, 0, mp->m_sb.sb_blocksize);
 		block->bb_magic = cpu_to_be32(XFS_ABTB_MAGIC);
@@ -278,8 +279,9 @@ xfs_growfs_data_private(
 		 * CNT btree root block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-			XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
-			BTOBB(mp->m_sb.sb_blocksize), 0);
+				 XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
+				 BTOBB(mp->m_sb.sb_blocksize),
+				 XBF_LOCK | XBF_MAPPED);
 		block = XFS_BUF_TO_BLOCK(bp);
 		memset(block, 0, mp->m_sb.sb_blocksize);
 		block->bb_magic = cpu_to_be32(XFS_ABTC_MAGIC);
@@ -300,8 +302,9 @@ xfs_growfs_data_private(
 		 * INO btree root block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-			XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
-			BTOBB(mp->m_sb.sb_blocksize), 0);
+				 XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
+				 BTOBB(mp->m_sb.sb_blocksize),
+				 XBF_LOCK | XBF_MAPPED);
 		block = XFS_BUF_TO_BLOCK(bp);
 		memset(block, 0, mp->m_sb.sb_blocksize);
 		block->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
@@ -611,7 +614,7 @@ 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);

fs/xfs/xfs_ialloc.c

@@ -425,7 +425,7 @@ xfs_ialloc_ag_alloc(
 	return 0;
 }
 
-STATIC_INLINE xfs_agnumber_t
+STATIC xfs_agnumber_t
 xfs_ialloc_next_ag(
 	xfs_mount_t	*mp)
 {

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;
@@ -290,7 +293,7 @@ xfs_iget_cache_miss(
 	struct xfs_inode	**ipp,
 	xfs_daddr_t		bno,
 	int			flags,
-	int			lock_flags) __releases(pag->pag_ici_lock)
+	int			lock_flags)
 {
 	struct xfs_inode	*ip;
 	int			error;

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,

fs/xfs/xfs_log_recover.c

@@ -2206,6 +2206,7 @@ xlog_recover_do_buffer_trans(
 	xfs_daddr_t		blkno;
 	int			len;
 	ushort			flags;
+	uint			buf_flags;
 
 	buf_f = (xfs_buf_log_format_t *)item->ri_buf[0].i_addr;
 
@@ -2246,12 +2247,11 @@ xlog_recover_do_buffer_trans(
 	}
 
 	mp = log->l_mp;
-	if (flags & XFS_BLI_INODE_BUF) {
-		bp = xfs_buf_read_flags(mp->m_ddev_targp, blkno, len,
-								XFS_BUF_LOCK);
-	} else {
-		bp = xfs_buf_read(mp->m_ddev_targp, blkno, len, 0);
-	}
+	buf_flags = XFS_BUF_LOCK;
+	if (!(flags & XFS_BLI_INODE_BUF))
+		buf_flags |= XFS_BUF_MAPPED;
+
+	bp = xfs_buf_read(mp->m_ddev_targp, blkno, len, buf_flags);
 	if (XFS_BUF_ISERROR(bp)) {
 		xfs_ioerror_alert("xlog_recover_do..(read#1)", log->l_mp,
 				  bp, blkno);
@@ -2350,8 +2350,8 @@ xlog_recover_do_inode_trans(
 		goto error;
 	}
 
-	bp = xfs_buf_read_flags(mp->m_ddev_targp, in_f->ilf_blkno,
-				in_f->ilf_len, XFS_BUF_LOCK);
+	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len,
+			  XFS_BUF_LOCK);
 	if (XFS_BUF_ISERROR(bp)) {
 		xfs_ioerror_alert("xlog_recover_do..(read#2)", mp,
 				  bp, in_f->ilf_blkno);
@@ -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,

fs/xfs/xfs_mount.c

@@ -583,8 +583,8 @@ xfs_readsb(xfs_mount_t *mp, int flags)
 	sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
 	extra_flags = XFS_BUF_LOCK | XFS_BUF_MANAGE | XFS_BUF_MAPPED;
 
-	bp = xfs_buf_read_flags(mp->m_ddev_targp, XFS_SB_DADDR,
-				BTOBB(sector_size), extra_flags);
+	bp = xfs_buf_read(mp->m_ddev_targp, XFS_SB_DADDR, BTOBB(sector_size),
+			  extra_flags);
 	if (!bp || XFS_BUF_ISERROR(bp)) {
 		xfs_fs_mount_cmn_err(flags, "SB read failed");
 		error = bp ? XFS_BUF_GETERROR(bp) : ENOMEM;
@@ -624,8 +624,8 @@ xfs_readsb(xfs_mount_t *mp, int flags)
 		XFS_BUF_UNMANAGE(bp);
 		xfs_buf_relse(bp);
 		sector_size = mp->m_sb.sb_sectsize;
-		bp = xfs_buf_read_flags(mp->m_ddev_targp, XFS_SB_DADDR,
-					BTOBB(sector_size), extra_flags);
+		bp = xfs_buf_read(mp->m_ddev_targp, XFS_SB_DADDR,
+				  BTOBB(sector_size), extra_flags);
 		if (!bp || XFS_BUF_ISERROR(bp)) {
 			xfs_fs_mount_cmn_err(flags, "SB re-read failed");
 			error = bp ? XFS_BUF_GETERROR(bp) : ENOMEM;
@@ -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) {
@@ -2123,7 +2123,7 @@ xfs_icsb_destroy_counters(
 	mutex_destroy(&mp->m_icsb_mutex);
 }
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_lock_cntr(
 	xfs_icsb_cnts_t	*icsbp)
 {
@@ -2132,7 +2132,7 @@ xfs_icsb_lock_cntr(
 	}
 }
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_unlock_cntr(
 	xfs_icsb_cnts_t	*icsbp)
 {
@@ -2140,7 +2140,7 @@ xfs_icsb_unlock_cntr(
 }
 
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_lock_all_counters(
 	xfs_mount_t	*mp)
 {
@@ -2153,7 +2153,7 @@ xfs_icsb_lock_all_counters(
 	}
 }
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_unlock_all_counters(
 	xfs_mount_t	*mp)
 {

fs/xfs/xfs_mount.h

@@ -93,6 +93,9 @@ typedef struct xfs_dmops {
 	xfs_send_unmount_t	xfs_send_unmount;
 } xfs_dmops_t;
 
+#define XFS_DMAPI_UNMOUNT_FLAGS(mp) \
+	(((mp)->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ? 0 : DM_FLAGS_UNWANTED)
+
 #define XFS_SEND_DATA(mp, ev,ip,off,len,fl,lock) \
 	(*(mp)->m_dm_ops->xfs_send_data)(ev,ip,off,len,fl,lock)
 #define XFS_SEND_MMAP(mp, vma,fl) \
@@ -101,12 +104,24 @@ typedef struct xfs_dmops {
 	(*(mp)->m_dm_ops->xfs_send_destroy)(ip,right)
 #define XFS_SEND_NAMESP(mp, ev,b1,r1,b2,r2,n1,n2,mode,rval,fl) \
 	(*(mp)->m_dm_ops->xfs_send_namesp)(ev,NULL,b1,r1,b2,r2,n1,n2,mode,rval,fl)
-#define XFS_SEND_PREUNMOUNT(mp,b1,r1,b2,r2,n1,n2,mode,rval,fl) \
-	(*(mp)->m_dm_ops->xfs_send_namesp)(DM_EVENT_PREUNMOUNT,mp,b1,r1,b2,r2,n1,n2,mode,rval,fl)
 #define XFS_SEND_MOUNT(mp,right,path,name) \
 	(*(mp)->m_dm_ops->xfs_send_mount)(mp,right,path,name)
-#define XFS_SEND_UNMOUNT(mp, ip,right,mode,rval,fl) \
-	(*(mp)->m_dm_ops->xfs_send_unmount)(mp,ip,right,mode,rval,fl)
+#define XFS_SEND_PREUNMOUNT(mp) \
+do { \
+	if (mp->m_flags & XFS_MOUNT_DMAPI) { \
+		(*(mp)->m_dm_ops->xfs_send_namesp)(DM_EVENT_PREUNMOUNT, mp, \
+			(mp)->m_rootip, DM_RIGHT_NULL, \
+			(mp)->m_rootip, DM_RIGHT_NULL, \
+			NULL, NULL, 0, 0, XFS_DMAPI_UNMOUNT_FLAGS(mp)); \
+	} \
+} while (0)
+#define XFS_SEND_UNMOUNT(mp) \
+do { \
+	if (mp->m_flags & XFS_MOUNT_DMAPI) { \
+		(*(mp)->m_dm_ops->xfs_send_unmount)(mp, (mp)->m_rootip, \
+			DM_RIGHT_NULL, 0, 0, XFS_DMAPI_UNMOUNT_FLAGS(mp)); \
+	} \
+} while (0)
 
 
 #ifdef HAVE_PERCPU_SB
@@ -387,13 +402,13 @@ xfs_put_perag(struct xfs_mount *mp, xfs_perag_t *pag)
  * Per-cpu superblock locking functions
  */
 #ifdef HAVE_PERCPU_SB
-STATIC_INLINE void
+static inline void
 xfs_icsb_lock(xfs_mount_t *mp)
 {
 	mutex_lock(&mp->m_icsb_mutex);
 }
 
-STATIC_INLINE void
+static inline void
 xfs_icsb_unlock(xfs_mount_t *mp)
 {
 	mutex_unlock(&mp->m_icsb_mutex);

fs/xfs/xfs_rw.c

@@ -277,10 +277,10 @@ xfs_read_buf(
 	xfs_buf_t	 *bp;
 	int		 error;
 
-	if (flags)
-		bp = xfs_buf_read_flags(target, blkno, len, flags);
-	else
-		bp = xfs_buf_read(target, blkno, len, flags);
+	if (!flags)
+		flags = XBF_LOCK | XBF_MAPPED;
+
+	bp = xfs_buf_read(target, blkno, len, flags);
 	if (!bp)
 		return XFS_ERROR(EIO);
 	error = XFS_BUF_GETERROR(bp);
@@ -336,3 +336,25 @@ xfs_bwrite(
 	}
 	return (error);
 }
+
+/*
+ * helper function to extract extent size hint from inode
+ */
+xfs_extlen_t
+xfs_get_extsz_hint(
+	struct xfs_inode	*ip)
+{
+	xfs_extlen_t		extsz;
+
+	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
+		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
+				? ip->i_d.di_extsize
+				: ip->i_mount->m_sb.sb_rextsize;
+		ASSERT(extsz);
+	} else {
+		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
+				? ip->i_d.di_extsize : 0;
+	}
+
+	return extsz;
+}

fs/xfs/xfs_rw.h

@@ -36,34 +36,6 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
 		 XFS_FSB_TO_DADDR((ip)->i_mount, (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
-xfs_get_extsz_hint(
-	xfs_inode_t	*ip)
-{
-	xfs_extlen_t	extsz;
-
-	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
-		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
-				? ip->i_d.di_extsize
-				: ip->i_mount->m_sb.sb_rextsize;
-		ASSERT(extsz);
-	} else {
-		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
-				? ip->i_d.di_extsize : 0;
-	}
-	return extsz;
-}
-
 /*
  * Prototypes for functions in xfs_rw.c.
  */
@@ -76,5 +48,6 @@ extern int xfs_read_buf(struct xfs_mount *mp, xfs_buftarg_t *btp,
 			struct xfs_buf **bpp);
 extern void xfs_ioerror_alert(char *func, struct xfs_mount *mp,
 				xfs_buf_t *bp, xfs_daddr_t blkno);
+extern xfs_extlen_t xfs_get_extsz_hint(struct xfs_inode *ip);
 
 #endif /* __XFS_RW_H__ */

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;

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);

fs/xfs/xfs_trans_buf.c

@@ -79,11 +79,8 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	/*
 	 * Default to a normal get_buf() call if the tp is NULL.
 	 */
-	if (tp == NULL) {
-		bp = xfs_buf_get_flags(target_dev, blkno, len,
-							flags | BUF_BUSY);
-		return(bp);
-	}
+	if (tp == NULL)
+		return xfs_buf_get(target_dev, blkno, len, flags | BUF_BUSY);
 
 	/*
 	 * If we find the buffer in the cache with this transaction
@@ -129,7 +126,7 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	 * easily deadlock with our current transaction as well as cause
 	 * us to run out of stack space.
 	 */
-	bp = xfs_buf_get_flags(target_dev, blkno, len, flags | BUF_BUSY);
+	bp = xfs_buf_get(target_dev, blkno, len, flags | BUF_BUSY);
 	if (bp == NULL) {
 		return NULL;
 	}
@@ -302,7 +299,7 @@ xfs_trans_read_buf(
 	 * Default to a normal get_buf() call if the tp is NULL.
 	 */
 	if (tp == NULL) {
-		bp = xfs_buf_read_flags(target, blkno, len, flags | BUF_BUSY);
+		bp = xfs_buf_read(target, blkno, len, flags | BUF_BUSY);
 		if (!bp)
 			return (flags & XFS_BUF_TRYLOCK) ?
 					EAGAIN : XFS_ERROR(ENOMEM);
@@ -398,7 +395,7 @@ xfs_trans_read_buf(
 	 * easily deadlock with our current transaction as well as cause
 	 * us to run out of stack space.
 	 */
-	bp = xfs_buf_read_flags(target, blkno, len, flags | BUF_BUSY);
+	bp = xfs_buf_read(target, blkno, len, flags | BUF_BUSY);
 	if (bp == NULL) {
 		*bpp = NULL;
 		return 0;

fs/xfs/xfs_vnodeops.c

@@ -538,9 +538,8 @@ xfs_readlink_bmap(
 		d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
 		byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
 
-		bp = xfs_buf_read_flags(mp->m_ddev_targp, d, BTOBB(byte_cnt),
-					XBF_LOCK | XBF_MAPPED |
-					XBF_DONT_BLOCK);
+		bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt),
+				  XBF_LOCK | XBF_MAPPED | XBF_DONT_BLOCK);
 		error = XFS_BUF_GETERROR(bp);
 		if (error) {
 			xfs_ioerror_alert("xfs_readlink",
@@ -708,6 +707,11 @@ xfs_fsync(
 	return error;
 }
 
+/*
+ * 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
@@ -726,7 +730,6 @@ 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 +771,19 @@ 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 +820,7 @@ 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 +1120,17 @@ 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 +1201,7 @@ 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 +2473,6 @@ 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.

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,