hugetlb: use free_pool_huge_page() to return unused surplus pages
authorLee Schermerhorn <lee.schermerhorn@hp.com>
Tue, 22 Sep 2009 00:01:23 +0000 (17:01 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 22 Sep 2009 14:17:26 +0000 (07:17 -0700)
Use the [modified] free_pool_huge_page() function to return unused
surplus pages.  This will help keep huge pages balanced across nodes
between freeing of unused surplus pages and freeing of persistent huge
pages [from set_max_huge_pages] by using the same node id "cursor". It
also eliminates some code duplication.

Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Nishanth Aravamudan <nacc@us.ibm.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@canonical.com>
Cc: Eric Whitney <eric.whitney@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/hugetlb.c

index 38dab55..f10cc27 100644 (file)
@@ -687,7 +687,7 @@ static int hstate_next_node_to_free(struct hstate *h)
  * balanced over allowed nodes.
  * Called with hugetlb_lock locked.
  */
-static int free_pool_huge_page(struct hstate *h)
+static int free_pool_huge_page(struct hstate *h, bool acct_surplus)
 {
        int start_nid;
        int next_nid;
@@ -697,13 +697,22 @@ static int free_pool_huge_page(struct hstate *h)
        next_nid = start_nid;
 
        do {
-               if (!list_empty(&h->hugepage_freelists[next_nid])) {
+               /*
+                * If we're returning unused surplus pages, only examine
+                * nodes with surplus pages.
+                */
+               if ((!acct_surplus || h->surplus_huge_pages_node[next_nid]) &&
+                   !list_empty(&h->hugepage_freelists[next_nid])) {
                        struct page *page =
                                list_entry(h->hugepage_freelists[next_nid].next,
                                          struct page, lru);
                        list_del(&page->lru);
                        h->free_huge_pages--;
                        h->free_huge_pages_node[next_nid]--;
+                       if (acct_surplus) {
+                               h->surplus_huge_pages--;
+                               h->surplus_huge_pages_node[next_nid]--;
+                       }
                        update_and_free_page(h, page);
                        ret = 1;
                }
@@ -884,22 +893,13 @@ free:
  * When releasing a hugetlb pool reservation, any surplus pages that were
  * allocated to satisfy the reservation must be explicitly freed if they were
  * never used.
+ * Called with hugetlb_lock held.
  */
 static void return_unused_surplus_pages(struct hstate *h,
                                        unsigned long unused_resv_pages)
 {
-       static int nid = -1;
-       struct page *page;
        unsigned long nr_pages;
 
-       /*
-        * We want to release as many surplus pages as possible, spread
-        * evenly across all nodes. Iterate across all nodes until we
-        * can no longer free unreserved surplus pages. This occurs when
-        * the nodes with surplus pages have no free pages.
-        */
-       unsigned long remaining_iterations = nr_online_nodes;
-
        /* Uncommit the reservation */
        h->resv_huge_pages -= unused_resv_pages;
 
@@ -909,26 +909,17 @@ static void return_unused_surplus_pages(struct hstate *h,
 
        nr_pages = min(unused_resv_pages, h->surplus_huge_pages);
 
-       while (remaining_iterations-- && nr_pages) {
-               nid = next_node(nid, node_online_map);
-               if (nid == MAX_NUMNODES)
-                       nid = first_node(node_online_map);
-
-               if (!h->surplus_huge_pages_node[nid])
-                       continue;
-
-               if (!list_empty(&h->hugepage_freelists[nid])) {
-                       page = list_entry(h->hugepage_freelists[nid].next,
-                                         struct page, lru);
-                       list_del(&page->lru);
-                       update_and_free_page(h, page);
-                       h->free_huge_pages--;
-                       h->free_huge_pages_node[nid]--;
-                       h->surplus_huge_pages--;
-                       h->surplus_huge_pages_node[nid]--;
-                       nr_pages--;
-                       remaining_iterations = nr_online_nodes;
-               }
+       /*
+        * We want to release as many surplus pages as possible, spread
+        * evenly across all nodes. Iterate across all nodes until we
+        * can no longer free unreserved surplus pages. This occurs when
+        * the nodes with surplus pages have no free pages.
+        * free_pool_huge_page() will balance the the frees across the
+        * on-line nodes for us and will handle the hstate accounting.
+        */
+       while (nr_pages--) {
+               if (!free_pool_huge_page(h, 1))
+                       break;
        }
 }
 
@@ -1268,7 +1259,7 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count)
        min_count = max(count, min_count);
        try_to_free_low(h, min_count);
        while (min_count < persistent_huge_pages(h)) {
-               if (!free_pool_huge_page(h))
+               if (!free_pool_huge_page(h, 0))
                        break;
        }
        while (count < persistent_huge_pages(h)) {