When thread A calls kill_f2fs_super, thread A may submit a bio to the function
iput(sbi->node_inode);
Then thread A enters the process D state, and thread B may also call
kill_f2fs_super and execute the code sbi->node_inode = NULL;
Now that the bio submitted by thread A is complete, it calls f2fs_write_end_io
and may trigger null-ptr-deref in NODE_MAPPING.
Thread A Thread B
IRQ context
- f2fs_put_super
- iput(sbi->node_inode)
- iput_final
- write_inode_now
- writeback_single_inode
- __writeback_single_inode
- filemap_fdatawait
- filemap_fdatawait_range
- __kcfi_typeid_free_transhuge_page
- __filemap_fdatawait_range
- wait_on_page_writeback
- folio_wait_writeback
- folio_wait_bit
- folio_wait_bit_common
- io_schedule
- f2fs_put_super
- sbi->node_inode = NULL
- __handle_irq_event_percpu
- ufs_qcom_mcq_esi_handler
- ufshcd_mcq_poll_cqe_nolock
- ufshcd_compl_one_cqe
- scsi_done
- scsi_done_internal
- blk_mq_complete_request
- scsi_complete
- scsi_finish_command
- scsi_io_completion
- scsi_end_request
- blk_update_request
- bio_endio
- f2fs_write_end_io
- NODE_MAPPING(sbi)
Signed-off-by: sunshijie <sunshi...@xiaomi.com>
---
fs/f2fs/super.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index a4bc26dfdb1a..adffc9b80a9c 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -1618,7 +1618,6 @@ static void f2fs_put_super(struct super_block *sb)
f2fs_release_ino_entry(sbi, true);
f2fs_leave_shrinker(sbi);
- mutex_unlock(&sbi->umount_mutex);
/* our cp_error case, we can wait for any writeback page */
f2fs_flush_merged_writes(sbi);
@@ -1648,6 +1647,8 @@ static void f2fs_put_super(struct super_block *sb)
iput(sbi->meta_inode);
sbi->meta_inode = NULL;
+ mutex_unlock(&sbi->umount_mutex);
+
/*
* iput() can update stat information, if f2fs_write_checkpoint()
* above failed with error.
--
2.34.1
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
