Btrfs: Add zlib compression support

[mirror_ubuntu-artful-kernel.git] / fs / btrfs / ordered-data.c

diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index e42fd233e04cdef068acb94df92303b1aa09dc75..b5745bb96d40ddfc34d7d3719c5c4ab5b04df790 100644 (file)
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -26,7 +26,6 @@
 #include "btrfs_inode.h"
 #include "extent_io.h"
 
-
 static u64 entry_end(struct btrfs_ordered_extent *entry)
 {
        if (entry->file_offset + entry->len < entry->file_offset)
@@ -34,6 +33,9 @@ static u64 entry_end(struct btrfs_ordered_extent *entry)
        return entry->file_offset + entry->len;
 }
 
+/* returns NULL if the insertion worked, or it returns the node it did find
+ * in the tree
+ */
 static struct rb_node *tree_insert(struct rb_root *root, u64 file_offset,
                                   struct rb_node *node)
 {
@@ -58,6 +60,10 @@ static struct rb_node *tree_insert(struct rb_root *root, u64 file_offset,
        return NULL;
 }
 
+/*
+ * look for a given offset in the tree, and if it can't be found return the
+ * first lesser offset
+ */
 static struct rb_node *__tree_search(struct rb_root *root, u64 file_offset,
                                     struct rb_node **prev_ret)
 {
@@ -108,6 +114,9 @@ static struct rb_node *__tree_search(struct rb_root *root, u64 file_offset,
        return NULL;
 }
 
+/*
+ * helper to check if a given offset is inside a given entry
+ */
 static int offset_in_entry(struct btrfs_ordered_extent *entry, u64 file_offset)
 {
        if (file_offset < entry->file_offset ||
@@ -116,6 +125,10 @@ static int offset_in_entry(struct btrfs_ordered_extent *entry, u64 file_offset)
        return 1;
 }
 
+/*
+ * look find the first ordered struct that has this offset, otherwise
+ * the first one less than this offset
+ */
 static inline struct rb_node *tree_search(struct btrfs_ordered_inode_tree *tree,
                                          u64 file_offset)
 {
@@ -152,7 +165,8 @@ static inline struct rb_node *tree_search(struct btrfs_ordered_inode_tree *tree,
  * inserted.
  */
 int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
-                            u64 start, u64 len)
+                            u64 start, u64 len, u64 disk_len, int nocow,
+                            int compressed)
 {
        struct btrfs_ordered_inode_tree *tree;
        struct rb_node *node;
@@ -167,20 +181,33 @@ int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
        entry->file_offset = file_offset;
        entry->start = start;
        entry->len = len;
+       entry->disk_len = disk_len;
+       entry->inode = inode;
+       if (nocow)
+               set_bit(BTRFS_ORDERED_NOCOW, &entry->flags);
+       if (compressed)
+               set_bit(BTRFS_ORDERED_COMPRESSED, &entry->flags);
+
        /* one ref for the tree */
        atomic_set(&entry->refs, 1);
        init_waitqueue_head(&entry->wait);
        INIT_LIST_HEAD(&entry->list);
+       INIT_LIST_HEAD(&entry->root_extent_list);
 
        node = tree_insert(&tree->tree, file_offset,
                           &entry->rb_node);
        if (node) {
-               entry = rb_entry(node, struct btrfs_ordered_extent, rb_node);
-               atomic_inc(&entry->refs);
+               printk("warning dup entry from add_ordered_extent\n");
+               BUG();
        }
        set_extent_ordered(&BTRFS_I(inode)->io_tree, file_offset,
                           entry_end(entry) - 1, GFP_NOFS);
 
+       spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+       list_add_tail(&entry->root_extent_list,
+                     &BTRFS_I(inode)->root->fs_info->ordered_extents);
+       spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+
        mutex_unlock(&tree->mutex);
        BUG_ON(node);
        return 0;
@@ -285,11 +312,67 @@ int btrfs_remove_ordered_extent(struct inode *inode,
        rb_erase(node, &tree->tree);
        tree->last = NULL;
        set_bit(BTRFS_ORDERED_COMPLETE, &entry->flags);
+
+       spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+       list_del_init(&entry->root_extent_list);
+       spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+
        mutex_unlock(&tree->mutex);
        wake_up(&entry->wait);
        return 0;
 }
 
+/*
+ * wait for all the ordered extents in a root.  This is done when balancing
+ * space between drives.
+ */
+int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only)
+{
+       struct list_head splice;
+       struct list_head *cur;
+       struct btrfs_ordered_extent *ordered;
+       struct inode *inode;
+
+       INIT_LIST_HEAD(&splice);
+
+       spin_lock(&root->fs_info->ordered_extent_lock);
+       list_splice_init(&root->fs_info->ordered_extents, &splice);
+       while (!list_empty(&splice)) {
+               cur = splice.next;
+               ordered = list_entry(cur, struct btrfs_ordered_extent,
+                                    root_extent_list);
+               if (nocow_only &&
+                   !test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags)) {
+                       list_move(&ordered->root_extent_list,
+                                 &root->fs_info->ordered_extents);
+                       cond_resched_lock(&root->fs_info->ordered_extent_lock);
+                       continue;
+               }
+
+               list_del_init(&ordered->root_extent_list);
+               atomic_inc(&ordered->refs);
+
+               /*
+                * the inode may be getting freed (in sys_unlink path).
+                */
+               inode = igrab(ordered->inode);
+
+               spin_unlock(&root->fs_info->ordered_extent_lock);
+
+               if (inode) {
+                       btrfs_start_ordered_extent(inode, ordered, 1);
+                       btrfs_put_ordered_extent(ordered);
+                       iput(inode);
+               } else {
+                       btrfs_put_ordered_extent(ordered);
+               }
+
+               spin_lock(&root->fs_info->ordered_extent_lock);
+       }
+       spin_unlock(&root->fs_info->ordered_extent_lock);
+       return 0;
+}
+
 /*
  * Used to start IO or wait for a given ordered extent to finish.
  *
@@ -310,15 +393,16 @@ void btrfs_start_ordered_extent(struct inode *inode,
         * for pdflush to find them
         */
        btrfs_fdatawrite_range(inode->i_mapping, start, end, WB_SYNC_NONE);
-       if (wait)
+       if (wait) {
                wait_event(entry->wait, test_bit(BTRFS_ORDERED_COMPLETE,
                                                 &entry->flags));
+       }
 }
 
 /*
  * Used to wait on ordered extents across a large range of bytes.
  */
-void btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
+int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 {
        u64 end;
        u64 orig_end;
@@ -372,6 +456,7 @@ again:
                       (unsigned long long)orig_end);
                goto again;
        }
+       return 0;
 }
 
 /*
@@ -496,7 +581,7 @@ int btrfs_ordered_update_i_size(struct inode *inode,
                 */
                test = rb_entry(node, struct btrfs_ordered_extent, rb_node);
                if (test->file_offset > entry_end(ordered)) {
-                       i_size_test = test->file_offset - 1;
+                       i_size_test = test->file_offset;
                }
        } else {
                i_size_test = i_size_read(inode);
@@ -509,7 +594,7 @@ int btrfs_ordered_update_i_size(struct inode *inode,
         * disk_i_size to the end of the region.
         */
        if (i_size_test > entry_end(ordered) &&
-           !test_range_bit(io_tree, entry_end(ordered), i_size_test,
+           !test_range_bit(io_tree, entry_end(ordered), i_size_test - 1,
                           EXTENT_DELALLOC, 0)) {
                new_i_size = min_t(u64, i_size_test, i_size_read(inode));
        }