btrfs: fix possible deadlock by clearing __GFP_FS flag
Using the GFP_HIGHUSER_MOVABLE flag to allocate the metadata's page may cause
deadlock.
Task1
open()
...
btrfs_search_slot()
...
btrfs_cow_block()
...
alloc_page()
wait for reclaiming
shrink_slab()
...
shrink_icache_memory()
...
btrfs_evict_inode()
...
btrfs_search_slot()
If the path is locked by task1, the deadlock happens.
So the btree's page cache is different with the file's page cache, it can not
allocate pages by GFP_HIGHUSER_MOVABLE flag, we must clear __GFP_FS flag in
GFP_HIGHUSER_MOVABLE flag.
Reported-by: Itaru Kitayama <kitayama@cl.bb4u.ne.jp>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index b3fc847..5cf3aa7 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1724,6 +1724,8 @@
goto fail_bdi;
}
+ fs_info->btree_inode->i_mapping->flags &= ~__GFP_FS;
+
INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
INIT_LIST_HEAD(&fs_info->trans_list);
INIT_LIST_HEAD(&fs_info->dead_roots);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 04babaf..0627418 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2536,6 +2536,8 @@
BTRFS_I(inode)->flags = btrfs_inode_flags(leaf, inode_item);
alloc_group_block = btrfs_inode_block_group(leaf, inode_item);
+ if (location.objectid == BTRFS_FREE_SPACE_OBJECTID)
+ inode->i_mapping->flags &= ~__GFP_FS;
/*
* try to precache a NULL acl entry for files that don't have
@@ -4084,7 +4086,6 @@
BTRFS_I(inode)->root = root;
memcpy(&BTRFS_I(inode)->location, location, sizeof(*location));
btrfs_read_locked_inode(inode);
-
inode_tree_add(inode);
unlock_new_inode(inode);
if (new)