Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux...
[firefly-linux-kernel-4.4.55.git] / fs / btrfs / disk-io.c
index e1fe74a2ce16e6a4e0b38129160f484e642c42fa..2936ca49b3b4af3a799f7585b74038d289ab8278 100644 (file)
@@ -44,6 +44,7 @@
 #include "free-space-cache.h"
 #include "inode-map.h"
 #include "check-integrity.h"
+#include "rcu-string.h"
 
 static struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
@@ -1153,7 +1154,6 @@ static void __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
        root->orphan_block_rsv = NULL;
 
        INIT_LIST_HEAD(&root->dirty_list);
-       INIT_LIST_HEAD(&root->orphan_list);
        INIT_LIST_HEAD(&root->root_list);
        spin_lock_init(&root->orphan_lock);
        spin_lock_init(&root->inode_lock);
@@ -1166,6 +1166,7 @@ static void __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
        atomic_set(&root->log_commit[0], 0);
        atomic_set(&root->log_commit[1], 0);
        atomic_set(&root->log_writers, 0);
+       atomic_set(&root->orphan_inodes, 0);
        root->log_batch = 0;
        root->log_transid = 0;
        root->last_log_commit = 0;
@@ -1252,7 +1253,7 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 
        leaf = btrfs_alloc_free_block(trans, root, root->leafsize, 0,
                                      BTRFS_TREE_LOG_OBJECTID, NULL,
-                                     0, 0, 0, 0);
+                                     0, 0, 0);
        if (IS_ERR(leaf)) {
                kfree(root);
                return ERR_CAST(leaf);
@@ -1914,11 +1915,14 @@ int open_ctree(struct super_block *sb,
        spin_lock_init(&fs_info->delayed_iput_lock);
        spin_lock_init(&fs_info->defrag_inodes_lock);
        spin_lock_init(&fs_info->free_chunk_lock);
+       spin_lock_init(&fs_info->tree_mod_seq_lock);
+       rwlock_init(&fs_info->tree_mod_log_lock);
        mutex_init(&fs_info->reloc_mutex);
 
        init_completion(&fs_info->kobj_unregister);
        INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
        INIT_LIST_HEAD(&fs_info->space_info);
+       INIT_LIST_HEAD(&fs_info->tree_mod_seq_list);
        btrfs_mapping_init(&fs_info->mapping_tree);
        btrfs_init_block_rsv(&fs_info->global_block_rsv);
        btrfs_init_block_rsv(&fs_info->delalloc_block_rsv);
@@ -1931,12 +1935,14 @@ int open_ctree(struct super_block *sb,
        atomic_set(&fs_info->async_submit_draining, 0);
        atomic_set(&fs_info->nr_async_bios, 0);
        atomic_set(&fs_info->defrag_running, 0);
+       atomic_set(&fs_info->tree_mod_seq, 0);
        fs_info->sb = sb;
        fs_info->max_inline = 8192 * 1024;
        fs_info->metadata_ratio = 0;
        fs_info->defrag_inodes = RB_ROOT;
        fs_info->trans_no_join = 0;
        fs_info->free_chunk_space = 0;
+       fs_info->tree_mod_log = RB_ROOT;
 
        /* readahead state */
        INIT_RADIX_TREE(&fs_info->reada_tree, GFP_NOFS & ~__GFP_WAIT);
@@ -2001,7 +2007,8 @@ int open_ctree(struct super_block *sb,
        BTRFS_I(fs_info->btree_inode)->root = tree_root;
        memset(&BTRFS_I(fs_info->btree_inode)->location, 0,
               sizeof(struct btrfs_key));
-       BTRFS_I(fs_info->btree_inode)->dummy_inode = 1;
+       set_bit(BTRFS_INODE_DUMMY,
+               &BTRFS_I(fs_info->btree_inode)->runtime_flags);
        insert_inode_hash(fs_info->btree_inode);
 
        spin_lock_init(&fs_info->block_group_cache_lock);
@@ -2112,7 +2119,7 @@ int open_ctree(struct super_block *sb,
 
        features = btrfs_super_incompat_flags(disk_super);
        features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
-       if (tree_root->fs_info->compress_type & BTRFS_COMPRESS_LZO)
+       if (tree_root->fs_info->compress_type == BTRFS_COMPRESS_LZO)
                features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
 
        /*
@@ -2347,12 +2354,24 @@ retry_root_backup:
                                  BTRFS_CSUM_TREE_OBJECTID, csum_root);
        if (ret)
                goto recovery_tree_root;
-
        csum_root->track_dirty = 1;
 
        fs_info->generation = generation;
        fs_info->last_trans_committed = generation;
 
+       ret = btrfs_recover_balance(fs_info);
+       if (ret) {
+               printk(KERN_WARNING "btrfs: failed to recover balance\n");
+               goto fail_block_groups;
+       }
+
+       ret = btrfs_init_dev_stats(fs_info);
+       if (ret) {
+               printk(KERN_ERR "btrfs: failed to init dev_stats: %d\n",
+                      ret);
+               goto fail_block_groups;
+       }
+
        ret = btrfs_init_space_info(fs_info);
        if (ret) {
                printk(KERN_ERR "Failed to initial space info: %d\n", ret);
@@ -2471,20 +2490,23 @@ retry_root_backup:
                goto fail_trans_kthread;
        }
 
-       if (!(sb->s_flags & MS_RDONLY)) {
-               down_read(&fs_info->cleanup_work_sem);
-               err = btrfs_orphan_cleanup(fs_info->fs_root);
-               if (!err)
-                       err = btrfs_orphan_cleanup(fs_info->tree_root);
-               up_read(&fs_info->cleanup_work_sem);
+       if (sb->s_flags & MS_RDONLY)
+               return 0;
 
-               if (!err)
-                       err = btrfs_recover_balance(fs_info->tree_root);
+       down_read(&fs_info->cleanup_work_sem);
+       if ((ret = btrfs_orphan_cleanup(fs_info->fs_root)) ||
+           (ret = btrfs_orphan_cleanup(fs_info->tree_root))) {
+               up_read(&fs_info->cleanup_work_sem);
+               close_ctree(tree_root);
+               return ret;
+       }
+       up_read(&fs_info->cleanup_work_sem);
 
-               if (err) {
-                       close_ctree(tree_root);
-                       return err;
-               }
+       ret = btrfs_resume_balance_async(fs_info);
+       if (ret) {
+               printk(KERN_WARNING "btrfs: failed to resume balance\n");
+               close_ctree(tree_root);
+               return ret;
        }
 
        return 0;
@@ -2556,18 +2578,20 @@ recovery_tree_root:
 
 static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate)
 {
-       char b[BDEVNAME_SIZE];
-
        if (uptodate) {
                set_buffer_uptodate(bh);
        } else {
-               printk_ratelimited(KERN_WARNING "lost page write due to "
-                                       "I/O error on %s\n",
-                                      bdevname(bh->b_bdev, b));
+               struct btrfs_device *device = (struct btrfs_device *)
+                       bh->b_private;
+
+               printk_ratelimited_in_rcu(KERN_WARNING "lost page write due to "
+                                         "I/O error on %s\n",
+                                         rcu_str_deref(device->name));
                /* note, we dont' set_buffer_write_io_error because we have
                 * our own ways of dealing with the IO errors
                 */
                clear_buffer_uptodate(bh);
+               btrfs_dev_stat_inc_and_print(device, BTRFS_DEV_STAT_WRITE_ERRS);
        }
        unlock_buffer(bh);
        put_bh(bh);
@@ -2682,6 +2706,7 @@ static int write_dev_supers(struct btrfs_device *device,
                        set_buffer_uptodate(bh);
                        lock_buffer(bh);
                        bh->b_end_io = btrfs_end_buffer_write_sync;
+                       bh->b_private = device;
                }
 
                /*
@@ -2734,12 +2759,15 @@ static int write_dev_flush(struct btrfs_device *device, int wait)
                wait_for_completion(&device->flush_wait);
 
                if (bio_flagged(bio, BIO_EOPNOTSUPP)) {
-                       printk("btrfs: disabling barriers on dev %s\n",
-                              device->name);
+                       printk_in_rcu("btrfs: disabling barriers on dev %s\n",
+                                     rcu_str_deref(device->name));
                        device->nobarriers = 1;
                }
                if (!bio_flagged(bio, BIO_UPTODATE)) {
                        ret = -EIO;
+                       if (!bio_flagged(bio, BIO_EOPNOTSUPP))
+                               btrfs_dev_stat_inc_and_print(device,
+                                       BTRFS_DEV_STAT_FLUSH_ERRS);
                }
 
                /* drop the reference from the wait == 0 run */
@@ -2902,19 +2930,6 @@ int write_ctree_super(struct btrfs_trans_handle *trans,
        return ret;
 }
 
-/* Kill all outstanding I/O */
-void btrfs_abort_devices(struct btrfs_root *root)
-{
-       struct list_head *head;
-       struct btrfs_device *dev;
-       mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
-       head = &root->fs_info->fs_devices->devices;
-       list_for_each_entry_rcu(dev, head, dev_list) {
-               blk_abort_queue(dev->bdev->bd_disk->queue);
-       }
-       mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
-}
-
 void btrfs_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root)
 {
        spin_lock(&fs_info->fs_roots_radix_lock);
@@ -3395,7 +3410,6 @@ int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 
        delayed_refs = &trans->delayed_refs;
 
-again:
        spin_lock(&delayed_refs->lock);
        if (delayed_refs->num_entries == 0) {
                spin_unlock(&delayed_refs->lock);
@@ -3403,31 +3417,37 @@ again:
                return ret;
        }
 
-       node = rb_first(&delayed_refs->root);
-       while (node) {
+       while ((node = rb_first(&delayed_refs->root)) != NULL) {
                ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
-               node = rb_next(node);
-
-               ref->in_tree = 0;
-               rb_erase(&ref->rb_node, &delayed_refs->root);
-               delayed_refs->num_entries--;
 
                atomic_set(&ref->refs, 1);
                if (btrfs_delayed_ref_is_head(ref)) {
                        struct btrfs_delayed_ref_head *head;
 
                        head = btrfs_delayed_node_to_head(ref);
-                       spin_unlock(&delayed_refs->lock);
-                       mutex_lock(&head->mutex);
+                       if (!mutex_trylock(&head->mutex)) {
+                               atomic_inc(&ref->refs);
+                               spin_unlock(&delayed_refs->lock);
+
+                               /* Need to wait for the delayed ref to run */
+                               mutex_lock(&head->mutex);
+                               mutex_unlock(&head->mutex);
+                               btrfs_put_delayed_ref(ref);
+
+                               spin_lock(&delayed_refs->lock);
+                               continue;
+                       }
+
                        kfree(head->extent_op);
                        delayed_refs->num_heads--;
                        if (list_empty(&head->cluster))
                                delayed_refs->num_heads_ready--;
                        list_del_init(&head->cluster);
-                       mutex_unlock(&head->mutex);
-                       btrfs_put_delayed_ref(ref);
-                       goto again;
                }
+               ref->in_tree = 0;
+               rb_erase(&ref->rb_node, &delayed_refs->root);
+               delayed_refs->num_entries--;
+
                spin_unlock(&delayed_refs->lock);
                btrfs_put_delayed_ref(ref);
 
@@ -3515,11 +3535,9 @@ static int btrfs_destroy_marked_extents(struct btrfs_root *root,
                             &(&BTRFS_I(page->mapping->host)->io_tree)->buffer,
                                               offset >> PAGE_CACHE_SHIFT);
                        spin_unlock(&dirty_pages->buffer_lock);
-                       if (eb) {
+                       if (eb)
                                ret = test_and_clear_bit(EXTENT_BUFFER_DIRTY,
                                                         &eb->bflags);
-                               atomic_set(&eb->refs, 1);
-                       }
                        if (PageWriteback(page))
                                end_page_writeback(page);
 
@@ -3533,8 +3551,8 @@ static int btrfs_destroy_marked_extents(struct btrfs_root *root,
                                spin_unlock_irq(&page->mapping->tree_lock);
                        }
 
-                       page->mapping->a_ops->invalidatepage(page, 0);
                        unlock_page(page);
+                       page_cache_release(page);
                }
        }
 
@@ -3548,8 +3566,10 @@ static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
        u64 start;
        u64 end;
        int ret;
+       bool loop = true;
 
        unpin = pinned_extents;
+again:
        while (1) {
                ret = find_first_extent_bit(unpin, 0, &start, &end,
                                            EXTENT_DIRTY);
@@ -3567,6 +3587,15 @@ static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
                cond_resched();
        }
 
+       if (loop) {
+               if (unpin == &root->fs_info->freed_extents[0])
+                       unpin = &root->fs_info->freed_extents[1];
+               else
+                       unpin = &root->fs_info->freed_extents[0];
+               loop = false;
+               goto again;
+       }
+
        return 0;
 }
 
@@ -3580,21 +3609,23 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
        /* FIXME: cleanup wait for commit */
        cur_trans->in_commit = 1;
        cur_trans->blocked = 1;
-       if (waitqueue_active(&root->fs_info->transaction_blocked_wait))
-               wake_up(&root->fs_info->transaction_blocked_wait);
+       wake_up(&root->fs_info->transaction_blocked_wait);
 
        cur_trans->blocked = 0;
-       if (waitqueue_active(&root->fs_info->transaction_wait))
-               wake_up(&root->fs_info->transaction_wait);
+       wake_up(&root->fs_info->transaction_wait);
 
        cur_trans->commit_done = 1;
-       if (waitqueue_active(&cur_trans->commit_wait))
-               wake_up(&cur_trans->commit_wait);
+       wake_up(&cur_trans->commit_wait);
+
+       btrfs_destroy_delayed_inodes(root);
+       btrfs_assert_delayed_root_empty(root);
 
        btrfs_destroy_pending_snapshots(cur_trans);
 
        btrfs_destroy_marked_extents(root, &cur_trans->dirty_pages,
                                     EXTENT_DIRTY);
+       btrfs_destroy_pinned_extent(root,
+                                   root->fs_info->pinned_extents);
 
        /*
        memset(cur_trans, 0, sizeof(*cur_trans));
@@ -3643,6 +3674,9 @@ int btrfs_cleanup_transaction(struct btrfs_root *root)
                if (waitqueue_active(&t->commit_wait))
                        wake_up(&t->commit_wait);
 
+               btrfs_destroy_delayed_inodes(root);
+               btrfs_assert_delayed_root_empty(root);
+
                btrfs_destroy_pending_snapshots(t);
 
                btrfs_destroy_delalloc_inodes(root);
@@ -3671,17 +3705,6 @@ int btrfs_cleanup_transaction(struct btrfs_root *root)
        return 0;
 }
 
-static int btree_writepage_io_failed_hook(struct bio *bio, struct page *page,
-                                         u64 start, u64 end,
-                                         struct extent_state *state)
-{
-       struct super_block *sb = page->mapping->host->i_sb;
-       struct btrfs_fs_info *fs_info = btrfs_sb(sb);
-       btrfs_error(fs_info, -EIO,
-                   "Error occured while writing out btree at %llu", start);
-       return -EIO;
-}
-
 static struct extent_io_ops btree_extent_io_ops = {
        .write_cache_pages_lock_hook = btree_lock_page_hook,
        .readpage_end_io_hook = btree_readpage_end_io_hook,
@@ -3689,5 +3712,4 @@ static struct extent_io_ops btree_extent_io_ops = {
        .submit_bio_hook = btree_submit_bio_hook,
        /* note we're sharing with inode.c for the merge bio hook */
        .merge_bio_hook = btrfs_merge_bio_hook,
-       .writepage_io_failed_hook = btree_writepage_io_failed_hook,
 };