1 files changed, 136 insertions, 67 deletions

diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 0037427d8a9d..11d2e9cea09e 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -24,6 +24,7 @@
 #include "free-space-cache.h"
 #include "transaction.h"
 #include "disk-io.h"
+#include "extent_io.h"
 
 #define BITS_PER_BITMAP         (PAGE_CACHE_SIZE * 8)
 #define MAX_CACHE_BYTES_PER_GIG (32 * 1024)
@@ -81,6 +82,8 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
                 return ERR_PTR(-ENOENT);
         }
 
+        inode->i_mapping->flags &= ~__GFP_FS;
+
         spin_lock(&block_group->lock);
         if (!root->fs_info->closing) {
                 block_group->inode = igrab(inode);
@@ -222,6 +225,7 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
         u64 num_entries;
         u64 num_bitmaps;
         u64 generation;
+        u64 used = btrfs_block_group_used(&block_group->item);
         u32 cur_crc = ~(u32)0;
         pgoff_t index = 0;
         unsigned long first_page_offset;
@@ -467,6 +471,17 @@ next:
                 index++;
         }
 
+        spin_lock(&block_group->tree_lock);
+        if (block_group->free_space != (block_group->key.offset - used -
+                                        block_group->bytes_super)) {
+                spin_unlock(&block_group->tree_lock);
+                printk(KERN_ERR "block group %llu has an wrong amount of free "
+                       "space\n", block_group->key.objectid);
+                ret = 0;
+                goto free_cache;
+        }
+        spin_unlock(&block_group->tree_lock);
+
         ret = 1;
 out:
         kfree(checksums);
@@ -493,18 +508,23 @@ int btrfs_write_out_cache(struct btrfs_root *root,
         struct inode *inode;
         struct rb_node *node;
         struct list_head *pos, *n;
+        struct page **pages;
         struct page *page;
         struct extent_state *cached_state = NULL;
+        struct btrfs_free_cluster *cluster = NULL;
+        struct extent_io_tree *unpin = NULL;
         struct list_head bitmap_list;
         struct btrfs_key key;
+        u64 start, end, len;
         u64 bytes = 0;
         u32 *crc, *checksums;
-        pgoff_t index = 0, last_index = 0;
         unsigned long first_page_offset;
-        int num_checksums;
+        int index = 0, num_pages = 0;
         int entries = 0;
         int bitmaps = 0;
         int ret = 0;
+        bool next_page = false;
+        bool out_of_space = false;
 
         root = root->fs_info->tree_root;
 
@@ -532,24 +552,43 @@ int btrfs_write_out_cache(struct btrfs_root *root,
                 return 0;
         }
 
-        last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
+        num_pages = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
+                PAGE_CACHE_SHIFT;
         filemap_write_and_wait(inode->i_mapping);
         btrfs_wait_ordered_range(inode, inode->i_size &
                                  ~(root->sectorsize - 1), (u64)-1);
 
         /* We need a checksum per page. */
-        num_checksums = i_size_read(inode) / PAGE_CACHE_SIZE;
-        crc = checksums  = kzalloc(sizeof(u32) * num_checksums, GFP_NOFS);
+        crc = checksums = kzalloc(sizeof(u32) * num_pages, GFP_NOFS);
         if (!crc) {
                 iput(inode);
                 return 0;
         }
 
+        pages = kzalloc(sizeof(struct page *) * num_pages, GFP_NOFS);
+        if (!pages) {
+                kfree(crc);
+                iput(inode);
+                return 0;
+        }
+
         /* Since the first page has all of our checksums and our generation we
          * need to calculate the offset into the page that we can start writing
          * our entries.
          */
-        first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64);
+        first_page_offset = (sizeof(u32) * num_pages) + sizeof(u64);
+
+        /* Get the cluster for this block_group if it exists */
+        if (!list_empty(&block_group->cluster_list))
+                cluster = list_entry(block_group->cluster_list.next,
+                                     struct btrfs_free_cluster,
+                                     block_group_list);
+
+        /*
+         * We shouldn't have switched the pinned extents yet so this is the
+         * right one
+         */
+        unpin = root->fs_info->pinned_extents;
 
         /*
          * Lock all pages first so we can lock the extent safely.
@@ -559,20 +598,18 @@ int btrfs_write_out_cache(struct btrfs_root *root,
          * after find_get_page at this point.  Just putting this here so people
          * know and don't freak out.
          */
-        while (index <= last_index) {
+        while (index < num_pages) {
                 page = grab_cache_page(inode->i_mapping, index);
                 if (!page) {
-                        pgoff_t i = 0;
+                        int i;
 
-                        while (i < index) {
-                                page = find_get_page(inode->i_mapping, i);
-                                unlock_page(page);
-                                page_cache_release(page);
-                                page_cache_release(page);
-                                i++;
+                        for (i = 0; i < num_pages; i++) {
+                                unlock_page(pages[i]);
+                                page_cache_release(pages[i]);
                         }
                         goto out_free;
                 }
+                pages[index] = page;
                 index++;
         }
 
@@ -580,6 +617,12 @@ int btrfs_write_out_cache(struct btrfs_root *root,
         lock_extent_bits(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1,
                          0, &cached_state, GFP_NOFS);
 
+        /*
+         * When searching for pinned extents, we need to start at our start
+         * offset.
+         */
+        start = block_group->key.objectid;
+
         /* Write out the extent entries */
         do {
                 struct btrfs_free_space_entry *entry;
@@ -587,18 +630,25 @@ int btrfs_write_out_cache(struct btrfs_root *root,
                 unsigned long offset = 0;
                 unsigned long start_offset = 0;
 
+                next_page = false;
+
                 if (index == 0) {
                         start_offset = first_page_offset;
                         offset = start_offset;
                 }
 
-                page = find_get_page(inode->i_mapping, index);
+                if (index >= num_pages) {
+                        out_of_space = true;
+                        break;
+                }
+
+                page = pages[index];
 
                 addr = kmap(page);
                 entry = addr + start_offset;
 
                 memset(addr, 0, PAGE_CACHE_SIZE);
-                while (1) {
+                while (node && !next_page) {
                         struct btrfs_free_space *e;
 
                         e = rb_entry(node, struct btrfs_free_space, offset_index);
@@ -614,12 +664,49 @@ int btrfs_write_out_cache(struct btrfs_root *root,
                                 entry->type = BTRFS_FREE_SPACE_EXTENT;
                         }
                         node = rb_next(node);
-                        if (!node)
-                                break;
+                        if (!node && cluster) {
+                                node = rb_first(&cluster->root);
+                                cluster = NULL;
+                        }
                         offset += sizeof(struct btrfs_free_space_entry);
                         if (offset + sizeof(struct btrfs_free_space_entry) >=
                             PAGE_CACHE_SIZE)
+                                next_page = true;
+                        entry++;
+                }
+
+                /*
+                 * We want to add any pinned extents to our free space cache
+                 * so we don't leak the space
+                 */
+                while (!next_page && (start < block_group->key.objectid +
+                                      block_group->key.offset)) {
+                        ret = find_first_extent_bit(unpin, start, &start, &end,
+                                                    EXTENT_DIRTY);
+                        if (ret) {
+                                ret = 0;
                                 break;
+                        }
+
+                        /* This pinned extent is out of our range */
+                        if (start >= block_group->key.objectid +
+                            block_group->key.offset)
+                                break;
+
+                        len = block_group->key.objectid +
+                                block_group->key.offset - start;
+                        len = min(len, end + 1 - start);
+
+                        entries++;
+                        entry->offset = cpu_to_le64(start);
+                        entry->bytes = cpu_to_le64(len);
+                        entry->type = BTRFS_FREE_SPACE_EXTENT;
+
+                        start = end + 1;
+                        offset += sizeof(struct btrfs_free_space_entry);
+                        if (offset + sizeof(struct btrfs_free_space_entry) >=
+                            PAGE_CACHE_SIZE)
+                                next_page = true;
                         entry++;
                 }
                 *crc = ~(u32)0;
@@ -632,25 +719,8 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 
                 bytes += PAGE_CACHE_SIZE;
 
-                ClearPageChecked(page);
-                set_page_extent_mapped(page);
-                SetPageUptodate(page);
-                set_page_dirty(page);
-
-                /*
-                 * We need to release our reference we got for grab_cache_page,
-                 * except for the first page which will hold our checksums, we
-                 * do that below.
-                 */
-                if (index != 0) {
-                        unlock_page(page);
-                        page_cache_release(page);
-                }
-
-                page_cache_release(page);
-
                 index++;
-        } while (node);
+        } while (node || next_page);
 
         /* Write out the bitmaps */
         list_for_each_safe(pos, n, &bitmap_list) {
@@ -658,7 +728,11 @@ int btrfs_write_out_cache(struct btrfs_root *root,
                 struct btrfs_free_space *entry =
                         list_entry(pos, struct btrfs_free_space, list);
 
-                page = find_get_page(inode->i_mapping, index);
+                if (index >= num_pages) {
+                        out_of_space = true;
+                        break;
+                }
+                page = pages[index];
 
                 addr = kmap(page);
                 memcpy(addr, entry->bitmap, PAGE_CACHE_SIZE);
@@ -669,64 +743,58 @@ int btrfs_write_out_cache(struct btrfs_root *root,
                 crc++;
                 bytes += PAGE_CACHE_SIZE;
 
-                ClearPageChecked(page);
-                set_page_extent_mapped(page);
-                SetPageUptodate(page);
-                set_page_dirty(page);
-                unlock_page(page);
-                page_cache_release(page);
-                page_cache_release(page);
                 list_del_init(&entry->list);
                 index++;
         }
 
+        if (out_of_space) {
+                btrfs_drop_pages(pages, num_pages);
+                unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
+                                     i_size_read(inode) - 1, &cached_state,
+                                     GFP_NOFS);
+                ret = 0;
+                goto out_free;
+        }
+
         /* Zero out the rest of the pages just to make sure */
-        while (index <= last_index) {
+        while (index < num_pages) {
                 void *addr;
 
-                page = find_get_page(inode->i_mapping, index);
-
+                page = pages[index];
                 addr = kmap(page);
                 memset(addr, 0, PAGE_CACHE_SIZE);
                 kunmap(page);
-                ClearPageChecked(page);
-                set_page_extent_mapped(page);
-                SetPageUptodate(page);
-                set_page_dirty(page);
-                unlock_page(page);
-                page_cache_release(page);
-                page_cache_release(page);
                 bytes += PAGE_CACHE_SIZE;
                 index++;
         }
 
-        btrfs_set_extent_delalloc(inode, 0, bytes - 1, &cached_state);
-
         /* Write the checksums and trans id to the first page */
         {
                 void *addr;
                 u64 *gen;
 
-                page = find_get_page(inode->i_mapping, 0);
+                page = pages[0];
 
                 addr = kmap(page);
-                memcpy(addr, checksums, sizeof(u32) * num_checksums);
-                gen = addr + (sizeof(u32) * num_checksums);
+                memcpy(addr, checksums, sizeof(u32) * num_pages);
+                gen = addr + (sizeof(u32) * num_pages);
                 *gen = trans->transid;
                 kunmap(page);
-                ClearPageChecked(page);
-                set_page_extent_mapped(page);
-                SetPageUptodate(page);
-                set_page_dirty(page);
-                unlock_page(page);
-                page_cache_release(page);
-                page_cache_release(page);
         }
-        BTRFS_I(inode)->generation = trans->transid;
 
+        ret = btrfs_dirty_pages(root, inode, pages, num_pages, 0,
+                                            bytes, &cached_state);
+        btrfs_drop_pages(pages, num_pages);
         unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
                              i_size_read(inode) - 1, &cached_state, GFP_NOFS);
 
+        if (ret) {
+                ret = 0;
+                goto out_free;
+        }
+
+        BTRFS_I(inode)->generation = trans->transid;
+
         filemap_write_and_wait(inode->i_mapping);
 
         key.objectid = BTRFS_FREE_SPACE_OBJECTID;
@@ -777,6 +845,7 @@ out_free:
                 BTRFS_I(inode)->generation = 0;
         }
         kfree(checksums);
+        kfree(pages);
         btrfs_update_inode(trans, root, inode);
         iput(inode);
         return ret;