[PATCH 1/2] xfs: cleanup xfs_iunpin_wait/xfs_iunpin_nowait

[PATCH 1/2] xfs: cleanup xfs_iunpin_wait/xfs_iunpin_nowait

[Christoph Hellwig]

Remove the inode item pointer and ili_last_lsn checks in __xfs_iunpin_wait
as any pinned inode is guaranteed to have them valid.  After this the
xfs_iunpin_nowait case is nothing more than a xfs_log_force_lsn, as we
know that the caller has already checked the pincount.

Make xfs_iunpin_nowait the new low-level routine just doing the log force
and rewrite xfs_iunpin_wait around it.


Signed-off-by: Christoph Hellwig <hch@lst.de>

Index: xfs/fs/xfs/xfs_inode.c
===================================================================
--- xfs.orig/fs/xfs/xfs_inode.c	2010-02-17 10:47:39.141254207 +0100
+++ xfs/fs/xfs/xfs_inode.c	2010-02-17 10:55:24.505255043 +0100
@@ -2467,47 +2467,31 @@ xfs_iunpin(
 }
 
 /*
- * This is called to unpin an inode. It can be directed to wait or to return
- * immediately without waiting for the inode to be unpinned.  The caller must
- * have the inode locked in at least shared mode so that the buffer cannot be
- * subsequently pinned once someone is waiting for it to be unpinned.
+ * This is called to unpin an inode.  The caller must have the inode locked
+ * in at least shared mode so that the buffer cannot be subsequently pinned
+ * once someone is waiting for it to be unpinned.
  */
-STATIC void
-__xfs_iunpin_wait(
-	xfs_inode_t	*ip,
-	int		wait)
+static void
+xfs_iunpin_nowait(
+	struct xfs_inode	*ip)
 {
-	xfs_inode_log_item_t	*iip = ip->i_itemp;
-
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
-	if (atomic_read(&ip->i_pincount) == 0)
-		return;
 
 	/* Give the log a push to start the unpinning I/O */
-	if (iip && iip->ili_last_lsn)
-		xfs_log_force_lsn(ip->i_mount, iip->ili_last_lsn, 0);
-	else
-		xfs_log_force(ip->i_mount, 0);
+	xfs_log_force_lsn(ip->i_mount, ip->i_itemp->ili_last_lsn, 0);
 
-	if (wait)
-		wait_event(ip->i_ipin_wait, (atomic_read(&ip->i_pincount) == 0));
 }
 
 void
 xfs_iunpin_wait(
-	xfs_inode_t	*ip)
+	struct xfs_inode	*ip)
 {
-	__xfs_iunpin_wait(ip, 1);
-}
-
-static inline void
-xfs_iunpin_nowait(
-	xfs_inode_t	*ip)
-{
-	__xfs_iunpin_wait(ip, 0);
+	if (xfs_ipincount(ip)) {
+		xfs_iunpin_nowait(ip);
+		wait_event(ip->i_ipin_wait, (xfs_ipincount(ip) == 0));
+	}
 }
 
-
 /*
  * xfs_iextents_copy()
  *

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH 1/2] xfs: cleanup xfs_iunpin_wait/xfs_iunpin_nowait

[Dave Chinner]

On Wed, Feb 17, 2010 at 02:43:56PM -0500, Christoph Hellwig wrote:
> Remove the inode item pointer and ili_last_lsn checks in __xfs_iunpin_wait
> as any pinned inode is guaranteed to have them valid.  After this the
> xfs_iunpin_nowait case is nothing more than a xfs_log_force_lsn, as we
> know that the caller has already checked the pincount.
> 
> Make xfs_iunpin_nowait the new low-level routine just doing the log force
> and rewrite xfs_iunpin_wait around it.
> 
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>

Nice cleanup.

Reviewed-by: Dave Chinner <david@fromorbit.com>

-- 
Dave Chinner
david@fromorbit.com

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs