2 files changed, 67 insertions, 7 deletions

diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index f433819611cb..a51a07c3a70c 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -144,6 +144,41 @@ restart:
         return last_error;
 }
 
+/*
+ * Select the next per-ag structure to iterate during the walk. The reclaim
+ * walk is optimised only to walk AGs with reclaimable inodes in them.
+ */
+static struct xfs_perag *
+xfs_inode_ag_iter_next_pag(
+        struct xfs_mount        *mp,
+        xfs_agnumber_t          *first,
+        int                     tag)
+{
+        struct xfs_perag        *pag = NULL;
+
+        if (tag == XFS_ICI_RECLAIM_TAG) {
+                int found;
+                int ref;
+
+                spin_lock(&mp->m_perag_lock);
+                found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
+                                (void **)&pag, *first, 1, tag);
+                if (found <= 0) {
+                        spin_unlock(&mp->m_perag_lock);
+                        return NULL;
+                }
+                *first = pag->pag_agno + 1;
+                /* open coded pag reference increment */
+                ref = atomic_inc_return(&pag->pag_ref);
+                spin_unlock(&mp->m_perag_lock);
+                trace_xfs_perag_get_reclaim(mp, pag->pag_agno, ref, _RET_IP_);
+        } else {
+                pag = xfs_perag_get(mp, *first);
+                (*first)++;
+        }
+        return pag;
+}
+
 int
 xfs_inode_ag_iterator(
         struct xfs_mount        *mp,
@@ -154,16 +189,15 @@ xfs_inode_ag_iterator(
         int                     exclusive,
         int                     *nr_to_scan)
 {
+        struct xfs_perag        *pag;
         int                     error = 0;
         int                     last_error = 0;
         xfs_agnumber_t          ag;
         int                     nr;
 
         nr = nr_to_scan ? *nr_to_scan : INT_MAX;
-        for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
-                struct xfs_perag        *pag;
-
-                pag = xfs_perag_get(mp, ag);
+        ag = 0;
+        while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag, tag))) {
                 error = xfs_inode_ag_walk(mp, pag, execute, flags, tag,
                                                 exclusive, &nr);
                 xfs_perag_put(pag);
@@ -640,6 +674,17 @@ __xfs_inode_set_reclaim_tag(
         radix_tree_tag_set(&pag->pag_ici_root,
                            XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino),
                            XFS_ICI_RECLAIM_TAG);
+
+        if (!pag->pag_ici_reclaimable) {
+                /* propagate the reclaim tag up into the perag radix tree */
+                spin_lock(&ip->i_mount->m_perag_lock);
+                radix_tree_tag_set(&ip->i_mount->m_perag_tree,
+                                XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+                                XFS_ICI_RECLAIM_TAG);
+                spin_unlock(&ip->i_mount->m_perag_lock);
+                trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno,
+                                                        -1, _RET_IP_);
+        }
         pag->pag_ici_reclaimable++;
 }
 
@@ -674,6 +719,16 @@ __xfs_inode_clear_reclaim_tag(
         radix_tree_tag_clear(&pag->pag_ici_root,
                         XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
         pag->pag_ici_reclaimable--;
+        if (!pag->pag_ici_reclaimable) {
+                /* clear the reclaim tag from the perag radix tree */
+                spin_lock(&ip->i_mount->m_perag_lock);
+                radix_tree_tag_clear(&ip->i_mount->m_perag_tree,
+                                XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+                                XFS_ICI_RECLAIM_TAG);
+                spin_unlock(&ip->i_mount->m_perag_lock);
+                trace_xfs_perag_clear_reclaim(ip->i_mount, pag->pag_agno,
+                                                        -1, _RET_IP_);
+        }
 }
 
 /*
@@ -838,7 +893,7 @@ xfs_reclaim_inode_shrink(
         struct xfs_mount *mp;
         struct xfs_perag *pag;
         xfs_agnumber_t  ag;
-        int             reclaimable = 0;
+        int             reclaimable;
 
         mp = container_of(shrink, struct xfs_mount, m_inode_shrink);
         if (nr_to_scan) {
@@ -852,8 +907,10 @@ xfs_reclaim_inode_shrink(
                         return -1;
        }
 
-        for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
-                pag = xfs_perag_get(mp, ag);
+        reclaimable = 0;
+        ag = 0;
+        while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag,
+                                        XFS_ICI_RECLAIM_TAG))) {
                 reclaimable += pag->pag_ici_reclaimable;
                 xfs_perag_put(pag);
         }
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index 73d5aa117384..302820690904 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -124,7 +124,10 @@ DEFINE_EVENT(xfs_perag_class, name,	\
                  unsigned long caller_ip),                                      \
         TP_ARGS(mp, agno, refcount, caller_ip))
 DEFINE_PERAG_REF_EVENT(xfs_perag_get);
+DEFINE_PERAG_REF_EVENT(xfs_perag_get_reclaim);
 DEFINE_PERAG_REF_EVENT(xfs_perag_put);
+DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim);
+DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim);
 
 TRACE_EVENT(xfs_attr_list_node_descend,
         TP_PROTO(struct xfs_attr_list_context *ctx,