Btrfs: try to free metadata pages when we free btree blocks

COW means we cycle though blocks fairly quickly, and once we
free an extent on disk, it doesn't make much sense to keep the pages around.

This commit tries to immediately free the page when we free the extent,
which lowers our memory footprint significantly.

Signed-off-by: Chris Mason <chris.mason@oracle.com>

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 0c482e0..f5e7cae 100644 (file)
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2384,6 +2384,10 @@ static int __free_extent(struct btrfs_trans_handle *trans,
                if (owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
                        ret = btrfs_del_csums(trans, root, bytenr, num_bytes);
                        BUG_ON(ret);
+               } else {
+                       invalidate_mapping_pages(info->btree_inode->i_mapping,
+                            bytenr >> PAGE_CACHE_SHIFT,
+                            (bytenr + num_bytes - 1) >> PAGE_CACHE_SHIFT);
                }
 
                ret = update_block_group(trans, root, bytenr, num_bytes, 0,