btrfs: defer adding raid type kobject until after chunk relocation
Any time the first block group of a new type is created, we add a new
kobject to sysfs to hold the attributes for that type. Kobject-internal
allocations always use GFP_KERNEL, making them prone to fs-reclaim races.
While it appears as if this can occur any time a block group is created,
the only times the first block group of a new type can be created in
memory is at mount and when we create the first new block group during
raid conversion.
This patch adds a new list to track pending kobject additions and then
handles them after we do chunk relocation. Between relocating the
target chunk (or forcing allocation of a new chunk in the case of data)
and removing the old chunk, we're in a safe place for fs-reclaim to
occur. We're holding the volume mutex, which is already held across
page faults, and the delete_unused_bgs_mutex, which will only stall
the cleaner thread.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index e10f74d..0b1f01d 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -9918,9 +9918,39 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
return 0;
}
+/* link_block_group will queue up kobjects to add when we're reclaim-safe */
+void btrfs_add_raid_kobjects(struct btrfs_fs_info *fs_info)
+{
+ struct btrfs_space_info *space_info;
+ struct raid_kobject *rkobj;
+ LIST_HEAD(list);
+ int index;
+ int ret = 0;
+
+ spin_lock(&fs_info->pending_raid_kobjs_lock);
+ list_splice_init(&fs_info->pending_raid_kobjs, &list);
+ spin_unlock(&fs_info->pending_raid_kobjs_lock);
+
+ list_for_each_entry(rkobj, &list, list) {
+ space_info = __find_space_info(fs_info, rkobj->flags);
+ index = btrfs_bg_flags_to_raid_index(rkobj->flags);
+
+ ret = kobject_add(&rkobj->kobj, &space_info->kobj,
+ "%s", get_raid_name(index));
+ if (ret) {
+ kobject_put(&rkobj->kobj);
+ break;
+ }
+ }
+ if (ret)
+ btrfs_warn(fs_info,
+ "failed to add kobject for block cache, ignoring");
+}
+
static void link_block_group(struct btrfs_block_group_cache *cache)
{
struct btrfs_space_info *space_info = cache->space_info;
+ struct btrfs_fs_info *fs_info = cache->fs_info;
int index = btrfs_bg_flags_to_raid_index(cache->flags);
bool first = false;
@@ -9931,27 +9961,20 @@ static void link_block_group(struct btrfs_block_group_cache *cache)
up_write(&space_info->groups_sem);
if (first) {
- struct raid_kobject *rkobj;
- int ret;
-
- rkobj = kzalloc(sizeof(*rkobj), GFP_NOFS);
- if (!rkobj)
- goto out_err;
- rkobj->raid_type = index;
- kobject_init(&rkobj->kobj, &btrfs_raid_ktype);
- ret = kobject_add(&rkobj->kobj, &space_info->kobj,
- "%s", get_raid_name(index));
- if (ret) {
- kobject_put(&rkobj->kobj);
- goto out_err;
+ struct raid_kobject *rkobj = kzalloc(sizeof(*rkobj), GFP_NOFS);
+ if (!rkobj) {
+ btrfs_warn(cache->fs_info,
+ "couldn't alloc memory for raid level kobject");
+ return;
}
+ rkobj->flags = cache->flags;
+ kobject_init(&rkobj->kobj, &btrfs_raid_ktype);
+
+ spin_lock(&fs_info->pending_raid_kobjs_lock);
+ list_add_tail(&rkobj->list, &fs_info->pending_raid_kobjs);
+ spin_unlock(&fs_info->pending_raid_kobjs_lock);
space_info->block_group_kobjs[index] = &rkobj->kobj;
}
-
- return;
-out_err:
- btrfs_warn(cache->fs_info,
- "failed to add kobject for block cache, ignoring");
}
static struct btrfs_block_group_cache *
@@ -10167,6 +10190,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
inc_block_group_ro(cache, 1);
}
+ btrfs_add_raid_kobjects(info);
init_global_block_rsv(info);
ret = 0;
error: