Btrfs: fix deadlock during allocating chunks

This deadlock comes from xfstests 251. We'll hold the chunk_mutex throughout the whole of a chunk allocation. But if we find that we've used up system chunk space, we need to allocate a new system chunk, but this will lead to a recursion of chunk allocation and end up with a deadlock on chunk_mutex. So instead we need to allocate the system chunk first if we find we're in ENOSPC. Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>

Btrfs: fix deadlock during allocating chunks
This deadlock comes from xfstests 251. We'll hold the chunk_mutex throughout the whole of a chunk allocation. But if we find that we've used up system chunk space, we need to allocate a new system chunk, but this will lead to a recursion of chunk allocation and end up with a deadlock on chunk_mutex. So instead we need to allocate the system chunk first if we find we're in ENOSPC. Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Liu Bo · Chris Mason
1 parent 2bcc0328c3
Showing 1 changed file with 50 additions and 0 deletions Side-by-side Diff
fs/btrfs/extent-tree.c
@@ -3445,6 +3445,50 @@
 	return 1;
 }
  
+static u64 get_system_chunk_thresh(struct btrfs_root *root, u64 type)
+{
+	u64 num_dev;
+
+	if (type & BTRFS_BLOCK_GROUP_RAID10 ||
+	    type & BTRFS_BLOCK_GROUP_RAID0)
+		num_dev = root->fs_info->fs_devices->rw_devices;
+	else if (type & BTRFS_BLOCK_GROUP_RAID1)
+		num_dev = 2;
+	else
+		num_dev = 1;	/* DUP or single */
+
+	/* metadata for updaing devices and chunk tree */
+	return btrfs_calc_trans_metadata_size(root, num_dev + 1);
+}
+
+static void check_system_chunk(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root, u64 type)
+{
+	struct btrfs_space_info *info;
+	u64 left;
+	u64 thresh;
+
+	info = __find_space_info(root->fs_info, BTRFS_BLOCK_GROUP_SYSTEM);
+	spin_lock(&info->lock);
+	left = info->total_bytes - info->bytes_used - info->bytes_pinned -
+		info->bytes_reserved - info->bytes_readonly;
+	spin_unlock(&info->lock);
+
+	thresh = get_system_chunk_thresh(root, type);
+	if (left < thresh && btrfs_test_opt(root, ENOSPC_DEBUG)) {
+		printk(KERN_INFO "left=%llu, need=%llu, flags=%llu\n",
+		       left, thresh, type);
+		dump_space_info(info, 0, 0);
+	}
+
+	if (left < thresh) {
+		u64 flags;
+
+		flags = btrfs_get_alloc_profile(root->fs_info->chunk_root, 0);
+		btrfs_alloc_chunk(trans, root, flags);
+	}
+}
+
 static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *extent_root, u64 alloc_bytes,
 			  u64 flags, int force)
@@ -3514,6 +3558,12 @@
 		      fs_info->metadata_ratio))
 			force_metadata_allocation(fs_info);
 	}
+
+	/*
+	 * Check if we have enough space in SYSTEM chunk because we may need
+	 * to update devices.
+	 */
+	check_system_chunk(trans, extent_root, flags);
  
 	ret = btrfs_alloc_chunk(trans, extent_root, flags);
 	if (ret < 0 && ret != -ENOSPC)
...	...	@@ -3445,6 +3445,50 @@
3445	3445	return 1;
3446	3446	}
3447	3447
	3448	+static u64 get_system_chunk_thresh(struct btrfs_root *root, u64 type)
	3449	+{
	3450	+ u64 num_dev;
	3451	+
	3452	+ if (type & BTRFS_BLOCK_GROUP_RAID10 \|\|
	3453	+ type & BTRFS_BLOCK_GROUP_RAID0)
	3454	+ num_dev = root->fs_info->fs_devices->rw_devices;
	3455	+ else if (type & BTRFS_BLOCK_GROUP_RAID1)
	3456	+ num_dev = 2;
	3457	+ else
	3458	+ num_dev = 1; /* DUP or single */
	3459	+
	3460	+ /* metadata for updaing devices and chunk tree */
	3461	+ return btrfs_calc_trans_metadata_size(root, num_dev + 1);
	3462	+}
	3463	+
	3464	+static void check_system_chunk(struct btrfs_trans_handle *trans,
	3465	+ struct btrfs_root *root, u64 type)
	3466	+{
	3467	+ struct btrfs_space_info *info;
	3468	+ u64 left;
	3469	+ u64 thresh;
	3470	+
	3471	+ info = __find_space_info(root->fs_info, BTRFS_BLOCK_GROUP_SYSTEM);
	3472	+ spin_lock(&info->lock);
	3473	+ left = info->total_bytes - info->bytes_used - info->bytes_pinned -
	3474	+ info->bytes_reserved - info->bytes_readonly;
	3475	+ spin_unlock(&info->lock);
	3476	+
	3477	+ thresh = get_system_chunk_thresh(root, type);
	3478	+ if (left < thresh && btrfs_test_opt(root, ENOSPC_DEBUG)) {
	3479	+ printk(KERN_INFO "left=%llu, need=%llu, flags=%llu\n",
	3480	+ left, thresh, type);
	3481	+ dump_space_info(info, 0, 0);
	3482	+ }
	3483	+
	3484	+ if (left < thresh) {
	3485	+ u64 flags;
	3486	+
	3487	+ flags = btrfs_get_alloc_profile(root->fs_info->chunk_root, 0);
	3488	+ btrfs_alloc_chunk(trans, root, flags);
	3489	+ }
	3490	+}
	3491	+
3448	3492	static int do_chunk_alloc(struct btrfs_trans_handle *trans,
3449	3493	struct btrfs_root *extent_root, u64 alloc_bytes,
3450	3494	u64 flags, int force)
...	...	@@ -3514,6 +3558,12 @@
3514	3558	fs_info->metadata_ratio))
3515	3559	force_metadata_allocation(fs_info);
3516	3560	}
	3561	+
	3562	+ /*
	3563	+ * Check if we have enough space in SYSTEM chunk because we may need
	3564	+ * to update devices.
	3565	+ */
	3566	+ check_system_chunk(trans, extent_root, flags);
3517	3567
3518	3568	ret = btrfs_alloc_chunk(trans, extent_root, flags);
3519	3569	if (ret < 0 && ret != -ENOSPC)