1 files changed, 168 insertions, 36 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 686331f22b15..ada4d24ed11b 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -59,6 +59,7 @@
 #include "backref.h"
 #include "hash.h"
 #include "props.h"
+#include "qgroup.h"
 
 struct btrfs_iget_args {
 	struct btrfs_key *location;
@@ -470,7 +471,7 @@ again:
 	 */
 	if (inode_need_compress(inode)) {
 		WARN_ON(pages);
-		pages = kzalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
+		pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS);
 		if (!pages) {
 			/* just bail out to the uncompressed code */
 			goto cont;
@@ -752,7 +753,6 @@ retry:
 			}
 			goto out_free;
 		}
-
 		/*
 		 * here we're doing allocation and writeback of the
 		 * compressed pages
@@ -3110,6 +3110,8 @@ void btrfs_run_delayed_iputs(struct btrfs_root *root)
 	if (empty)
 		return;
 
+	down_read(&fs_info->delayed_iput_sem);
+
 	spin_lock(&fs_info->delayed_iput_lock);
 	list_splice_init(&fs_info->delayed_iputs, &list);
 	spin_unlock(&fs_info->delayed_iput_lock);
@@ -3120,6 +3122,8 @@ void btrfs_run_delayed_iputs(struct btrfs_root *root)
 		iput(delayed->inode);
 		kfree(delayed);
 	}
+
+	up_read(&root->fs_info->delayed_iput_sem);
 }
 
 /*
@@ -4016,16 +4020,16 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
 {
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_trans_handle *trans;
-	struct inode *inode = dentry->d_inode;
+	struct inode *inode = d_inode(dentry);
 	int ret;
 
 	trans = __unlink_start_trans(dir);
 	if (IS_ERR(trans))
 		return PTR_ERR(trans);
 
-	btrfs_record_unlink_dir(trans, dir, dentry->d_inode, 0);
+	btrfs_record_unlink_dir(trans, dir, d_inode(dentry), 0);
 
-	ret = btrfs_unlink_inode(trans, root, dir, dentry->d_inode,
+	ret = btrfs_unlink_inode(trans, root, dir, d_inode(dentry),
 				 dentry->d_name.name, dentry->d_name.len);
 	if (ret)
 		goto out;
@@ -4124,7 +4128,7 @@ out:
 
 static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 {
-	struct inode *inode = dentry->d_inode;
+	struct inode *inode = d_inode(dentry);
 	int err = 0;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_trans_handle *trans;
@@ -4151,7 +4155,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 		goto out;
 
 	/* now the directory is empty */
-	err = btrfs_unlink_inode(trans, root, dir, dentry->d_inode,
+	err = btrfs_unlink_inode(trans, root, dir, d_inode(dentry),
 				 dentry->d_name.name, dentry->d_name.len);
 	if (!err)
 		btrfs_i_size_write(inode, 0);
@@ -4162,6 +4166,21 @@ out:
 	return err;
 }
 
+static int truncate_space_check(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root,
+				u64 bytes_deleted)
+{
+	int ret;
+
+	bytes_deleted = btrfs_csum_bytes_to_leaves(root, bytes_deleted);
+	ret = btrfs_block_rsv_add(root, &root->fs_info->trans_block_rsv,
+				  bytes_deleted, BTRFS_RESERVE_NO_FLUSH);
+	if (!ret)
+		trans->bytes_reserved += bytes_deleted;
+	return ret;
+
+}
+
 /*
  * this can truncate away extent items, csum items and directory items.
  * It starts at a high offset and removes keys until it can't find
@@ -4197,9 +4216,21 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 	int ret;
 	int err = 0;
 	u64 ino = btrfs_ino(inode);
+	u64 bytes_deleted = 0;
+	bool be_nice = 0;
+	bool should_throttle = 0;
+	bool should_end = 0;
 
 	BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY);
 
+	/*
+	 * for non-free space inodes and ref cows, we want to back off from
+	 * time to time
+	 */
+	if (!btrfs_is_free_space_inode(inode) &&
+	    test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+		be_nice = 1;
+
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -4229,6 +4260,19 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 	key.type = (u8)-1;
 
 search_again:
+	/*
+	 * with a 16K leaf size and 128MB extents, you can actually queue
+	 * up a huge file in a single leaf.  Most of the time that
+	 * bytes_deleted is > 0, it will be huge by the time we get here
+	 */
+	if (be_nice && bytes_deleted > 32 * 1024 * 1024) {
+		if (btrfs_should_end_transaction(trans, root)) {
+			err = -EAGAIN;
+			goto error;
+		}
+	}
+
+
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 	if (ret < 0) {
@@ -4371,22 +4415,39 @@ delete:
 		} else {
 			break;
 		}
+		should_throttle = 0;
+
 		if (found_extent &&
 		    (test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
 		     root == root->fs_info->tree_root)) {
 			btrfs_set_path_blocking(path);
+			bytes_deleted += extent_num_bytes;
 			ret = btrfs_free_extent(trans, root, extent_start,
 						extent_num_bytes, 0,
 						btrfs_header_owner(leaf),
 						ino, extent_offset, 0);
 			BUG_ON(ret);
+			if (btrfs_should_throttle_delayed_refs(trans, root))
+				btrfs_async_run_delayed_refs(root,
+					trans->delayed_ref_updates * 2, 0);
+			if (be_nice) {
+				if (truncate_space_check(trans, root,
+							 extent_num_bytes)) {
+					should_end = 1;
+				}
+				if (btrfs_should_throttle_delayed_refs(trans,
+								       root)) {
+					should_throttle = 1;
+				}
+			}
 		}
 
 		if (found_type == BTRFS_INODE_ITEM_KEY)
 			break;
 
 		if (path->slots[0] == 0 ||
-		    path->slots[0] != pending_del_slot) {
+		    path->slots[0] != pending_del_slot ||
+		    should_throttle || should_end) {
 			if (pending_del_nr) {
 				ret = btrfs_del_items(trans, root, path,
 						pending_del_slot,
@@ -4399,6 +4460,23 @@ delete:
 				pending_del_nr = 0;
 			}
 			btrfs_release_path(path);
+			if (should_throttle) {
+				unsigned long updates = trans->delayed_ref_updates;
+				if (updates) {
+					trans->delayed_ref_updates = 0;
+					ret = btrfs_run_delayed_refs(trans, root, updates * 2);
+					if (ret && !err)
+						err = ret;
+				}
+			}
+			/*
+			 * if we failed to refill our space rsv, bail out
+			 * and let the transaction restart
+			 */
+			if (should_end) {
+				err = -EAGAIN;
+				goto error;
+			}
 			goto search_again;
 		} else {
 			path->slots[0]--;
@@ -4415,7 +4493,18 @@ error:
 	if (last_size != (u64)-1 &&
 	    root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
 		btrfs_ordered_update_i_size(inode, last_size, NULL);
+
 	btrfs_free_path(path);
+
+	if (be_nice && bytes_deleted > 32 * 1024 * 1024) {
+		unsigned long updates = trans->delayed_ref_updates;
+		if (updates) {
+			trans->delayed_ref_updates = 0;
+			ret = btrfs_run_delayed_refs(trans, root, updates * 2);
+			if (ret && !err)
+				err = ret;
+		}
+	}
 	return err;
 }
 
@@ -4826,7 +4915,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 
 static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
-	struct inode *inode = dentry->d_inode;
+	struct inode *inode = d_inode(dentry);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int err;
 
@@ -4924,6 +5013,7 @@ void btrfs_evict_inode(struct inode *inode)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_block_rsv *rsv, *global_rsv;
+	int steal_from_global = 0;
 	u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
 	int ret;
 
@@ -4991,9 +5081,20 @@ void btrfs_evict_inode(struct inode *inode)
 		 * hard as possible to get this to work.
 		 */
 		if (ret)
-			ret = btrfs_block_rsv_migrate(global_rsv, rsv, min_size);
+			steal_from_global++;
+		else
+			steal_from_global = 0;
+		ret = 0;
 
-		if (ret) {
+		/*
+		 * steal_from_global == 0: we reserved stuff, hooray!
+		 * steal_from_global == 1: we didn't reserve stuff, boo!
+		 * steal_from_global == 2: we've committed, still not a lot of
+		 * room but maybe we'll have room in the global reserve this
+		 * time.
+		 * steal_from_global == 3: abandon all hope!
+		 */
+		if (steal_from_global > 2) {
 			btrfs_warn(root->fs_info,
 				"Could not get space for a delete, will truncate on mount %d",
 				ret);
@@ -5009,10 +5110,40 @@ void btrfs_evict_inode(struct inode *inode)
 			goto no_delete;
 		}
 
+		/*
+		 * We can't just steal from the global reserve, we need tomake
+		 * sure there is room to do it, if not we need to commit and try
+		 * again.
+		 */
+		if (steal_from_global) {
+			if (!btrfs_check_space_for_delayed_refs(trans, root))
+				ret = btrfs_block_rsv_migrate(global_rsv, rsv,
+							      min_size);
+			else
+				ret = -ENOSPC;
+		}
+
+		/*
+		 * Couldn't steal from the global reserve, we have too much
+		 * pending stuff built up, commit the transaction and try it
+		 * again.
+		 */
+		if (ret) {
+			ret = btrfs_commit_transaction(trans, root);
+			if (ret) {
+				btrfs_orphan_del(NULL, inode);
+				btrfs_free_block_rsv(root, rsv);
+				goto no_delete;
+			}
+			continue;
+		} else {
+			steal_from_global = 0;
+		}
+
 		trans->block_rsv = rsv;
 
 		ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
-		if (ret != -ENOSPC)
+		if (ret != -ENOSPC && ret != -EAGAIN)
 			break;
 
 		trans->block_rsv = &root->fs_info->trans_block_rsv;
@@ -5416,10 +5547,10 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 static int btrfs_dentry_delete(const struct dentry *dentry)
 {
 	struct btrfs_root *root;
-	struct inode *inode = dentry->d_inode;
+	struct inode *inode = d_inode(dentry);
 
 	if (!inode && !IS_ROOT(dentry))
-		inode = dentry->d_parent->d_inode;
+		inode = d_inode(dentry->d_parent);
 
 	if (inode) {
 		root = BTRFS_I(inode)->root;
@@ -6226,7 +6357,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 {
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
-	struct inode *inode = old_dentry->d_inode;
+	struct inode *inode = d_inode(old_dentry);
 	u64 index;
 	int err;
 	int drop_inode = 0;
@@ -8081,7 +8212,7 @@ free_ordered:
 	bio_endio(dio_bio, ret);
 }
 
-static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *iocb,
+static ssize_t check_direct_IO(struct btrfs_root *root, struct kiocb *iocb,
 			const struct iov_iter *iter, loff_t offset)
 {
 	int seg;
@@ -8096,7 +8227,7 @@ static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *io
 		goto out;
 
 	/* If this is a write we don't need to check anymore */
-	if (rw & WRITE)
+	if (iov_iter_rw(iter) == WRITE)
 		return 0;
 	/*
 	 * Check to make sure we don't have duplicate iov_base's in this
@@ -8114,8 +8245,8 @@ out:
 	return retval;
 }
 
-static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
-			struct iov_iter *iter, loff_t offset)
+static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
+			       loff_t offset)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_mapping->host;
@@ -8126,10 +8257,10 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 	bool relock = false;
 	ssize_t ret;
 
-	if (check_direct_IO(BTRFS_I(inode)->root, rw, iocb, iter, offset))
+	if (check_direct_IO(BTRFS_I(inode)->root, iocb, iter, offset))
 		return 0;
 
-	atomic_inc(&inode->i_dio_count);
+	inode_dio_begin(inode);
 	smp_mb__after_atomic();
 
 	/*
@@ -8144,7 +8275,7 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 		filemap_fdatawrite_range(inode->i_mapping, offset,
 					 offset + count - 1);
 
-	if (rw & WRITE) {
+	if (iov_iter_rw(iter) == WRITE) {
 		/*
 		 * If the write DIO is beyond the EOF, we need update
 		 * the isize, but it is protected by i_mutex. So we can
@@ -8169,16 +8300,16 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 		current->journal_info = &outstanding_extents;
 	} else if (test_bit(BTRFS_INODE_READDIO_NEED_LOCK,
 				     &BTRFS_I(inode)->runtime_flags)) {
-		inode_dio_done(inode);
+		inode_dio_end(inode);
 		flags = DIO_LOCKING | DIO_SKIP_HOLES;
 		wakeup = false;
 	}
 
-	ret = __blockdev_direct_IO(rw, iocb, inode,
-			BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev,
-			iter, offset, btrfs_get_blocks_direct, NULL,
-			btrfs_submit_direct, flags);
-	if (rw & WRITE) {
+	ret = __blockdev_direct_IO(iocb, inode,
+				   BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev,
+				   iter, offset, btrfs_get_blocks_direct, NULL,
+				   btrfs_submit_direct, flags);
+	if (iov_iter_rw(iter) == WRITE) {
 		current->journal_info = NULL;
 		if (ret < 0 && ret != -EIOCBQUEUED)
 			btrfs_delalloc_release_space(inode, count);
@@ -8188,7 +8319,7 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 	}
 out:
 	if (wakeup)
-		inode_dio_done(inode);
+		inode_dio_end(inode);
 	if (relock)
 		mutex_lock(&inode->i_mutex);
 
@@ -8581,7 +8712,7 @@ static int btrfs_truncate(struct inode *inode)
 		ret = btrfs_truncate_inode_items(trans, root, inode,
 						 inode->i_size,
 						 BTRFS_EXTENT_DATA_KEY);
-		if (ret != -ENOSPC) {
+		if (ret != -ENOSPC && ret != -EAGAIN) {
 			err = ret;
 			break;
 		}
@@ -8875,7 +9006,7 @@ static int btrfs_getattr(struct vfsmount *mnt,
 			 struct dentry *dentry, struct kstat *stat)
 {
 	u64 delalloc_bytes;
-	struct inode *inode = dentry->d_inode;
+	struct inode *inode = d_inode(dentry);
 	u32 blocksize = inode->i_sb->s_blocksize;
 
 	generic_fillattr(inode, stat);
@@ -8896,8 +9027,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(old_dir)->root;
 	struct btrfs_root *dest = BTRFS_I(new_dir)->root;
-	struct inode *new_inode = new_dentry->d_inode;
-	struct inode *old_inode = old_dentry->d_inode;
+	struct inode *new_inode = d_inode(new_dentry);
+	struct inode *old_inode = d_inode(old_dentry);
 	struct timespec ctime = CURRENT_TIME;
 	u64 index = 0;
 	u64 root_objectid;
@@ -9009,7 +9140,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 					old_dentry->d_name.len);
 	} else {
 		ret = __btrfs_unlink_inode(trans, root, old_dir,
-					old_dentry->d_inode,
+					d_inode(old_dentry),
 					old_dentry->d_name.name,
 					old_dentry->d_name.len);
 		if (!ret)
@@ -9033,12 +9164,12 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			BUG_ON(new_inode->i_nlink == 0);
 		} else {
 			ret = btrfs_unlink_inode(trans, dest, new_dir,
-						 new_dentry->d_inode,
+						 d_inode(new_dentry),
 						 new_dentry->d_name.name,
 						 new_dentry->d_name.len);
 		}
 		if (!ret && new_inode->i_nlink == 0)
-			ret = btrfs_orphan_add(trans, new_dentry->d_inode);
+			ret = btrfs_orphan_add(trans, d_inode(new_dentry));
 		if (ret) {
 			btrfs_abort_transaction(trans, root, ret);
 			goto out_fail;
@@ -9451,6 +9582,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 				btrfs_end_transaction(trans, root);
 			break;
 		}
+
 		btrfs_drop_extent_cache(inode, cur_offset,
 					cur_offset + ins.offset -1, 0);