f2fs: fix node_cnt race between extent node destroy and writeback

f2fs_destroy_extent_node() does not set FI_NO_EXTENT before clearing
extent nodes. When called from f2fs_drop_inode() with I_SYNC set,
concurrent kworker writeback can insert new extent nodes into the same
extent tree, racing with the destroy and triggering f2fs_bug_on() in
__destroy_extent_node(). The scenario is as follows:

drop inode                            writeback
 - iput
  - f2fs_drop_inode  // I_SYNC set
   - f2fs_destroy_extent_node
    - __destroy_extent_node
     - while (node_cnt) {
        write_lock(&et->lock)
        __free_extent_tree
        write_unlock(&et->lock)
                                       - __writeback_single_inode
                                        - f2fs_outplace_write_data
                                         - f2fs_update_read_extent_cache
                                          - __update_extent_tree_range
                                           // FI_NO_EXTENT not set,
                                           // insert new extent node
       } // node_cnt == 0, exit while
     - f2fs_bug_on(node_cnt)  // node_cnt > 0

Additionally, __update_extent_tree_range() only checks FI_NO_EXTENT for
EX_READ type, leaving EX_BLOCK_AGE updates completely unprotected.

This patch set FI_NO_EXTENT under et->lock in __destroy_extent_node(),
consistent with other callers (__update_extent_tree_range and
__drop_extent_tree) and check FI_NO_EXTENT for both EX_READ and
EX_BLOCK_AGE tree.

Fixes: 3fc5d5a182f6 ("f2fs: fix to shrink read extent node in batches")
Cc: stable@vger.kernel.org
Signed-off-by: Yongpeng Yang <yangyongpeng@xiaomi.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c
index 0ed84cc065a7eda66c50eae97d70e9ab93de31c3..87169fd29d897208776da9648989e46949fbf72b 100644 (file)
--- a/fs/f2fs/extent_cache.c
+++ b/fs/f2fs/extent_cache.c
@@ -119,9 +119,10 @@ static bool __may_extent_tree(struct inode *inode, enum extent_type type)
        if (!__init_may_extent_tree(inode, type))
                return false;
 
+       if (is_inode_flag_set(inode, FI_NO_EXTENT))
+               return false;
+
        if (type == EX_READ) {
-               if (is_inode_flag_set(inode, FI_NO_EXTENT))
-                       return false;
                if (is_inode_flag_set(inode, FI_COMPRESSED_FILE) &&
                                 !f2fs_sb_has_readonly(F2FS_I_SB(inode)))
                        return false;
@@ -644,6 +645,8 @@ static unsigned int __destroy_extent_node(struct inode *inode,
 
        while (atomic_read(&et->node_cnt)) {
                write_lock(&et->lock);
+               if (!is_inode_flag_set(inode, FI_NO_EXTENT))
+                       set_inode_flag(inode, FI_NO_EXTENT);
                node_cnt += __free_extent_tree(sbi, et, nr_shrink);
                write_unlock(&et->lock);
        }
@@ -688,12 +691,12 @@ static void __update_extent_tree_range(struct inode *inode,
 
        write_lock(&et->lock);
 
-       if (type == EX_READ) {
-               if (is_inode_flag_set(inode, FI_NO_EXTENT)) {
-                       write_unlock(&et->lock);
-                       return;
-               }
+       if (is_inode_flag_set(inode, FI_NO_EXTENT)) {
+               write_unlock(&et->lock);
+               return;
+       }
 
+       if (type == EX_READ) {
                prev = et->largest;
                dei.len = 0;