1 files changed, 128 insertions, 114 deletions

diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index c95f018d4a1e..c663aad880b9 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -152,7 +152,15 @@ struct list_head *btrfs_get_fs_uuids(void)
 	return &fs_uuids;
 }
 
-static struct btrfs_fs_devices *__alloc_fs_devices(void)
+/*
+ * alloc_fs_devices - allocate struct btrfs_fs_devices
+ * @fsid:	if not NULL, copy the uuid to fs_devices::fsid
+ *
+ * Return a pointer to a new struct btrfs_fs_devices on success, or ERR_PTR().
+ * The returned struct is not linked onto any lists and can be destroyed with
+ * kfree() right away.
+ */
+static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid)
 {
 	struct btrfs_fs_devices *fs_devs;
 
@@ -166,31 +174,8 @@ static struct btrfs_fs_devices *__alloc_fs_devices(void)
 	INIT_LIST_HEAD(&fs_devs->resized_devices);
 	INIT_LIST_HEAD(&fs_devs->alloc_list);
 	INIT_LIST_HEAD(&fs_devs->list);
-
-	return fs_devs;
-}
-
-/**
- * alloc_fs_devices - allocate struct btrfs_fs_devices
- * @fsid:	a pointer to UUID for this FS.  If NULL a new UUID is
- *		generated.
- *
- * Return: a pointer to a new &struct btrfs_fs_devices on success;
- * ERR_PTR() on error.  Returned struct is not linked onto any lists and
- * can be destroyed with kfree() right away.
- */
-static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid)
-{
-	struct btrfs_fs_devices *fs_devs;
-
-	fs_devs = __alloc_fs_devices();
-	if (IS_ERR(fs_devs))
-		return fs_devs;
-
 	if (fsid)
 		memcpy(fs_devs->fsid, fsid, BTRFS_FSID_SIZE);
-	else
-		generate_random_uuid(fs_devs->fsid);
 
 	return fs_devs;
 }
@@ -269,9 +254,17 @@ static struct btrfs_device *__alloc_device(void)
 	return dev;
 }
 
-static noinline struct btrfs_device *__find_device(struct list_head *head,
-						   u64 devid, u8 *uuid)
+/*
+ * Find a device specified by @devid or @uuid in the list of @fs_devices, or
+ * return NULL.
+ *
+ * If devid and uuid are both specified, the match must be exact, otherwise
+ * only devid is used.
+ */
+static struct btrfs_device *find_device(struct btrfs_fs_devices *fs_devices,
+		u64 devid, const u8 *uuid)
 {
+	struct list_head *head = &fs_devices->devices;
 	struct btrfs_device *dev;
 
 	list_for_each_entry(dev, head, dev_list) {
@@ -310,7 +303,7 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
 
 	if (flush)
 		filemap_write_and_wait((*bdev)->bd_inode->i_mapping);
-	ret = set_blocksize(*bdev, 4096);
+	ret = set_blocksize(*bdev, BTRFS_BDEV_BLOCKSIZE);
 	if (ret) {
 		blkdev_put(*bdev, flags);
 		goto error;
@@ -636,8 +629,8 @@ static noinline int device_list_add(const char *path,
 
 		device = NULL;
 	} else {
-		device = __find_device(&fs_devices->devices, devid,
-				       disk_super->dev_item.uuid);
+		device = find_device(fs_devices, devid,
+				disk_super->dev_item.uuid);
 	}
 
 	if (!device) {
@@ -1872,7 +1865,6 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	struct btrfs_fs_devices *cur_devices;
 	u64 num_devices;
 	int ret = 0;
-	bool clear_super = false;
 
 	mutex_lock(&uuid_mutex);
 
@@ -1908,7 +1900,6 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 		list_del_init(&device->dev_alloc_list);
 		device->fs_devices->rw_devices--;
 		mutex_unlock(&fs_info->chunk_mutex);
-		clear_super = true;
 	}
 
 	mutex_unlock(&uuid_mutex);
@@ -1987,9 +1978,6 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 		free_fs_devices(cur_devices);
 	}
 
-	fs_info->num_tolerated_disk_barrier_failures =
-		btrfs_calc_num_tolerated_disk_barrier_failures(fs_info);
-
 out:
 	mutex_unlock(&uuid_mutex);
 	return ret;
@@ -2202,7 +2190,7 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 	if (!fs_devices->seeding)
 		return -EINVAL;
 
-	seed_devices = __alloc_fs_devices();
+	seed_devices = alloc_fs_devices(NULL);
 	if (IS_ERR(seed_devices))
 		return PTR_ERR(seed_devices);
 
@@ -2407,7 +2395,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	device->is_tgtdev_for_dev_replace = 0;
 	device->mode = FMODE_EXCL;
 	device->dev_stats_valid = 1;
-	set_blocksize(device->bdev, 4096);
+	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
 
 	if (seeding_dev) {
 		sb->s_flags &= ~MS_RDONLY;
@@ -2487,8 +2475,6 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 				   "sysfs: failed to create fsid for sprout");
 	}
 
-	fs_info->num_tolerated_disk_barrier_failures =
-		btrfs_calc_num_tolerated_disk_barrier_failures(fs_info);
 	ret = btrfs_commit_transaction(trans);
 
 	if (seeding_dev) {
@@ -2612,7 +2598,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	device->is_tgtdev_for_dev_replace = 1;
 	device->mode = FMODE_EXCL;
 	device->dev_stats_valid = 1;
-	set_blocksize(device->bdev, 4096);
+	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
 	device->fs_devices = fs_info->fs_devices;
 	list_add(&device->dev_list, &fs_info->fs_devices->devices);
 	fs_info->fs_devices->num_devices++;
@@ -3312,7 +3298,6 @@ static int chunk_devid_filter(struct extent_buffer *leaf,
 /* [pstart, pend) */
 static int chunk_drange_filter(struct extent_buffer *leaf,
 			       struct btrfs_chunk *chunk,
-			       u64 chunk_offset,
 			       struct btrfs_balance_args *bargs)
 {
 	struct btrfs_stripe *stripe;
@@ -3439,7 +3424,7 @@ static int should_balance_chunk(struct btrfs_fs_info *fs_info,
 
 	/* drange filter, makes sense only with devid filter */
 	if ((bargs->flags & BTRFS_BALANCE_ARGS_DRANGE) &&
-	    chunk_drange_filter(leaf, chunk, chunk_offset, bargs)) {
+	    chunk_drange_filter(leaf, chunk, bargs)) {
 		return 0;
 	}
 
@@ -3898,13 +3883,6 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 			   meta_target, data_target);
 	}
 
-	if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
-		fs_info->num_tolerated_disk_barrier_failures = min(
-			btrfs_calc_num_tolerated_disk_barrier_failures(fs_info),
-			btrfs_get_num_tolerated_disk_barrier_failures(
-				bctl->sys.target));
-	}
-
 	ret = insert_balance_item(fs_info, bctl);
 	if (ret && ret != -EEXIST)
 		goto out;
@@ -3927,11 +3905,6 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 	mutex_lock(&fs_info->balance_mutex);
 	atomic_dec(&fs_info->balance_running);
 
-	if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
-		fs_info->num_tolerated_disk_barrier_failures =
-			btrfs_calc_num_tolerated_disk_barrier_failures(fs_info);
-	}
-
 	if (bargs) {
 		memset(bargs, 0, sizeof(*bargs));
 		update_ioctl_balance_args(fs_info, 0, bargs);
@@ -4127,7 +4100,6 @@ static int btrfs_uuid_scan_kthread(void *data)
 	struct btrfs_fs_info *fs_info = data;
 	struct btrfs_root *root = fs_info->tree_root;
 	struct btrfs_key key;
-	struct btrfs_key max_key;
 	struct btrfs_path *path = NULL;
 	int ret = 0;
 	struct extent_buffer *eb;
@@ -4146,10 +4118,6 @@ static int btrfs_uuid_scan_kthread(void *data)
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = 0;
 
-	max_key.objectid = (u64)-1;
-	max_key.type = BTRFS_ROOT_ITEM_KEY;
-	max_key.offset = (u64)-1;
-
 	while (1) {
 		ret = btrfs_search_forward(root, &key, path, 0);
 		if (ret) {
@@ -4601,12 +4569,6 @@ static int btrfs_cmp_device_info(const void *a, const void *b)
 	return 0;
 }
 
-static u32 find_raid56_stripe_len(u32 data_devices, u32 dev_stripe_target)
-{
-	/* TODO allow them to set a preferred stripe size */
-	return SZ_64K;
-}
-
 static void check_raid56_incompat_flag(struct btrfs_fs_info *info, u64 type)
 {
 	if (!(type & BTRFS_BLOCK_GROUP_RAID56_MASK))
@@ -4629,7 +4591,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *info = trans->fs_info;
 	struct btrfs_fs_devices *fs_devices = info->fs_devices;
-	struct list_head *cur;
+	struct btrfs_device *device;
 	struct map_lookup *map = NULL;
 	struct extent_map_tree *em_tree;
 	struct extent_map *em;
@@ -4649,7 +4611,6 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	u64 max_chunk_size;
 	u64 stripe_size;
 	u64 num_bytes;
-	u64 raid_stripe_len = BTRFS_STRIPE_LEN;
 	int ndevs;
 	int i;
 	int j;
@@ -4703,22 +4664,15 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	if (!devices_info)
 		return -ENOMEM;
 
-	cur = fs_devices->alloc_list.next;
-
 	/*
 	 * in the first pass through the devices list, we gather information
 	 * about the available holes on each device.
 	 */
 	ndevs = 0;
-	while (cur != &fs_devices->alloc_list) {
-		struct btrfs_device *device;
+	list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
 		u64 max_avail;
 		u64 dev_offset;
 
-		device = list_entry(cur, struct btrfs_device, dev_alloc_list);
-
-		cur = cur->next;
-
 		if (!device->writeable) {
 			WARN(1, KERN_ERR
 			       "BTRFS: read-only device in alloc_list\n");
@@ -4769,15 +4723,15 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	     btrfs_cmp_device_info, NULL);
 
 	/* round down to number of usable stripes */
-	ndevs -= ndevs % devs_increment;
+	ndevs = round_down(ndevs, devs_increment);
 
 	if (ndevs < devs_increment * sub_stripes || ndevs < devs_min) {
 		ret = -ENOSPC;
 		goto error;
 	}
 
-	if (devs_max && ndevs > devs_max)
-		ndevs = devs_max;
+	ndevs = min(ndevs, devs_max);
+
 	/*
 	 * the primary goal is to maximize the number of stripes, so use as many
 	 * devices as possible, even if the stripes are not maximum sized.
@@ -4791,16 +4745,11 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	 */
 	data_stripes = num_stripes / ncopies;
 
-	if (type & BTRFS_BLOCK_GROUP_RAID5) {
-		raid_stripe_len = find_raid56_stripe_len(ndevs - 1,
-							 info->stripesize);
+	if (type & BTRFS_BLOCK_GROUP_RAID5)
 		data_stripes = num_stripes - 1;
-	}
-	if (type & BTRFS_BLOCK_GROUP_RAID6) {
-		raid_stripe_len = find_raid56_stripe_len(ndevs - 2,
-							 info->stripesize);
+
+	if (type & BTRFS_BLOCK_GROUP_RAID6)
 		data_stripes = num_stripes - 2;
-	}
 
 	/*
 	 * Use the number of data stripes to figure out how big this chunk
@@ -4825,8 +4774,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	stripe_size = div_u64(stripe_size, dev_stripes);
 
 	/* align to BTRFS_STRIPE_LEN */
-	stripe_size = div64_u64(stripe_size, raid_stripe_len);
-	stripe_size *= raid_stripe_len;
+	stripe_size = round_down(stripe_size, BTRFS_STRIPE_LEN);
 
 	map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS);
 	if (!map) {
@@ -4844,9 +4792,9 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 		}
 	}
 	map->sector_size = info->sectorsize;
-	map->stripe_len = raid_stripe_len;
-	map->io_align = raid_stripe_len;
-	map->io_width = raid_stripe_len;
+	map->stripe_len = BTRFS_STRIPE_LEN;
+	map->io_align = BTRFS_STRIPE_LEN;
+	map->io_width = BTRFS_STRIPE_LEN;
 	map->type = type;
 	map->sub_stripes = sub_stripes;
 
@@ -5172,7 +5120,6 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
 }
 
 unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
-				    struct btrfs_mapping_tree *map_tree,
 				    u64 logical)
 {
 	struct extent_map *em;
@@ -5180,29 +5127,30 @@ unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
 	unsigned long len = fs_info->sectorsize;
 
 	em = get_chunk_map(fs_info, logical, len);
-	WARN_ON(IS_ERR(em));
 
-	map = em->map_lookup;
-	if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
-		len = map->stripe_len * nr_data_stripes(map);
-	free_extent_map(em);
+	if (!WARN_ON(IS_ERR(em))) {
+		map = em->map_lookup;
+		if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
+			len = map->stripe_len * nr_data_stripes(map);
+		free_extent_map(em);
+	}
 	return len;
 }
 
-int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
-			   u64 logical, u64 len, int mirror_num)
+int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
 {
 	struct extent_map *em;
 	struct map_lookup *map;
 	int ret = 0;
 
 	em = get_chunk_map(fs_info, logical, len);
-	WARN_ON(IS_ERR(em));
 
-	map = em->map_lookup;
-	if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
-		ret = 1;
-	free_extent_map(em);
+	if(!WARN_ON(IS_ERR(em))) {
+		map = em->map_lookup;
+		if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
+			ret = 1;
+		free_extent_map(em);
+	}
 	return ret;
 }
 
@@ -6295,8 +6243,7 @@ struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid,
 	while (cur_devices) {
 		if (!fsid ||
 		    !memcmp(cur_devices->fsid, fsid, BTRFS_UUID_SIZE)) {
-			device = __find_device(&cur_devices->devices,
-					       devid, uuid);
+			device = find_device(cur_devices, devid, uuid);
 			if (device)
 				return device;
 		}
@@ -6449,7 +6396,6 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 	struct extent_map *em;
 	u64 logical;
 	u64 length;
-	u64 stripe_len;
 	u64 devid;
 	u8 uuid[BTRFS_UUID_SIZE];
 	int num_stripes;
@@ -6458,7 +6404,6 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 
 	logical = key->offset;
 	length = btrfs_chunk_length(leaf, chunk);
-	stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
 	num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
 
 	ret = btrfs_check_chunk_valid(fs_info, leaf, chunk, logical);
@@ -6513,6 +6458,7 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 		if (!map->stripes[i].dev &&
 		    !btrfs_test_opt(fs_info, DEGRADED)) {
 			free_extent_map(em);
+			btrfs_report_missing_device(fs_info, devid, uuid);
 			return -EIO;
 		}
 		if (!map->stripes[i].dev) {
@@ -6523,8 +6469,7 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 				free_extent_map(em);
 				return -EIO;
 			}
-			btrfs_warn(fs_info, "devid %llu uuid %pU is missing",
-				   devid, uuid);
+			btrfs_report_missing_device(fs_info, devid, uuid);
 		}
 		map->stripes[i].dev->in_fs_metadata = 1;
 	}
@@ -6568,6 +6513,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 	int ret;
 
 	BUG_ON(!mutex_is_locked(&uuid_mutex));
+	ASSERT(fsid);
 
 	fs_devices = fs_info->fs_devices->seed;
 	while (fs_devices) {
@@ -6641,17 +6587,21 @@ static int read_one_dev(struct btrfs_fs_info *fs_info,
 
 	device = btrfs_find_device(fs_info, devid, dev_uuid, fs_uuid);
 	if (!device) {
-		if (!btrfs_test_opt(fs_info, DEGRADED))
+		if (!btrfs_test_opt(fs_info, DEGRADED)) {
+			btrfs_report_missing_device(fs_info, devid, dev_uuid);
 			return -EIO;
+		}
 
 		device = add_missing_dev(fs_devices, devid, dev_uuid);
 		if (!device)
 			return -ENOMEM;
-		btrfs_warn(fs_info, "devid %llu uuid %pU missing",
-				devid, dev_uuid);
+		btrfs_report_missing_device(fs_info, devid, dev_uuid);
 	} else {
-		if (!device->bdev && !btrfs_test_opt(fs_info, DEGRADED))
-			return -EIO;
+		if (!device->bdev) {
+			btrfs_report_missing_device(fs_info, devid, dev_uuid);
+			if (!btrfs_test_opt(fs_info, DEGRADED))
+				return -EIO;
+		}
 
 		if(!device->bdev && !device->missing) {
 			/*
@@ -6817,6 +6767,70 @@ out_short_read:
 	return -EIO;
 }
 
+void btrfs_report_missing_device(struct btrfs_fs_info *fs_info, u64 devid,
+				 u8 *uuid)
+{
+	btrfs_warn_rl(fs_info, "devid %llu uuid %pU is missing", devid, uuid);
+}
+
+/*
+ * Check if all chunks in the fs are OK for read-write degraded mount
+ *
+ * Return true if all chunks meet the minimal RW mount requirements.
+ * Return false if any chunk doesn't meet the minimal RW mount requirements.
+ */
+bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
+	struct extent_map *em;
+	u64 next_start = 0;
+	bool ret = true;
+
+	read_lock(&map_tree->map_tree.lock);
+	em = lookup_extent_mapping(&map_tree->map_tree, 0, (u64)-1);
+	read_unlock(&map_tree->map_tree.lock);
+	/* No chunk at all? Return false anyway */
+	if (!em) {
+		ret = false;
+		goto out;
+	}
+	while (em) {
+		struct map_lookup *map;
+		int missing = 0;
+		int max_tolerated;
+		int i;
+
+		map = em->map_lookup;
+		max_tolerated =
+			btrfs_get_num_tolerated_disk_barrier_failures(
+					map->type);
+		for (i = 0; i < map->num_stripes; i++) {
+			struct btrfs_device *dev = map->stripes[i].dev;
+
+			if (!dev || !dev->bdev || dev->missing ||
+			    dev->last_flush_error)
+				missing++;
+		}
+		if (missing > max_tolerated) {
+			btrfs_warn(fs_info,
+	"chunk %llu missing %d devices, max tolerance is %d for writeable mount",
+				   em->start, missing, max_tolerated);
+			free_extent_map(em);
+			ret = false;
+			goto out;
+		}
+		next_start = extent_map_end(em);
+		free_extent_map(em);
+
+		read_lock(&map_tree->map_tree.lock);
+		em = lookup_extent_mapping(&map_tree->map_tree, next_start,
+					   (u64)(-1) - next_start);
+		read_unlock(&map_tree->map_tree.lock);
+	}
+out:
+	return ret;
+}
+
 int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_root *root = fs_info->chunk_root;