[PATCH] xfs: use xfs_trans_getsb in xfs_sync_sb_buf

[Eric Sandeen <sandeen@redhat.com>]

Use xfs_trans_getsb rather than reaching right in for
mp->m_sb_bp; I think this is more correct, and it facilitates
building this libxfs code in userspace as well.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---

This passes the online label test ... tbh I'm not quite
sure if I'm unwinding/unlocking/putting the bp properly
here, though, so ... eyes & advice appreciated?

diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index 0bc0a40764e1..5f25aac04aa9 100644
--- a/fs/xfs/libxfs/xfs_sb.c
+++ b/fs/xfs/libxfs/xfs_sb.c
@@ -897,14 +897,16 @@ xfs_sync_sb_buf(
 	struct xfs_mount	*mp)
 {
 	struct xfs_trans	*tp;
+	struct xfs_buf		*bp;
 	int			error;
 
 	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_sb, 0, 0, 0, &tp);
 	if (error)
 		return error;
 
+	bp = xfs_trans_getsb(tp, mp, 0);
 	xfs_log_sb(tp);
-	xfs_trans_bhold(tp, mp->m_sb_bp);
+	xfs_trans_bhold(tp, bp);
 	xfs_trans_set_sync(tp);
 	error = xfs_trans_commit(tp);
 	if (error)
@@ -912,9 +914,9 @@ xfs_sync_sb_buf(
 	/*
 	 * write out the sb buffer to get the changes to disk
 	 */
-	error = xfs_bwrite(mp->m_sb_bp);
+	error = xfs_bwrite(bp);
 out:
-	xfs_buf_relse(mp->m_sb_bp);
+	xfs_buf_relse(bp);
 	return error;
 }