Commit fa91e4aa authored Jul 17, 2020 by Qu Wenruo Committed by David Sterba Jul 21, 2020

btrfs: qgroup: fix data leak caused by race between writeback and truncate



[BUG]
When running tests like generic/013 on test device with btrfs quota
enabled, it can normally lead to data leak, detected at unmount time:

  BTRFS warning (device dm-3): qgroup 0/5 has unreleased space, type 0 rsv 4096
  ------------[ cut here ]------------
  WARNING: CPU: 11 PID: 16386 at fs/btrfs/disk-io.c:4142 close_ctree+0x1dc/0x323 [btrfs]
  RIP: 0010:close_ctree+0x1dc/0x323 [btrfs]
  Call Trace:
   btrfs_put_super+0x15/0x17 [btrfs]
   generic_shutdown_super+0x72/0x110
   kill_anon_super+0x18/0x30
   btrfs_kill_super+0x17/0x30 [btrfs]
   deactivate_locked_super+0x3b/0xa0
   deactivate_super+0x40/0x50
   cleanup_mnt+0x135/0x190
   __cleanup_mnt+0x12/0x20
   task_work_run+0x64/0xb0
   __prepare_exit_to_usermode+0x1bc/0x1c0
   __syscall_return_slowpath+0x47/0x230
   do_syscall_64+0x64/0xb0
   entry_SYSCALL_64_after_hwframe+0x44/0xa9
  ---[ end trace caf08beafeca2392 ]---
  BTRFS error (device dm-3): qgroup reserved space leaked

[CAUSE]
In the offending case, the offending operations are:
2/6: writev f2X[269 1 0 0 0 0] [1006997,67,288] 0
2/7: truncate f2X[269 1 0 0 48 1026293] 18388 0

The following sequence of events could happen after the writev():
	CPU1 (writeback)		|		CPU2 (truncate)
-----------------------------------------------------------------
btrfs_writepages()			|
|- extent_write_cache_pages()		|
   |- Got page for 1003520		|
   |  1003520 is Dirty, no writeback	|
   |  So (!clear_page_dirty_for_io())   |
   |  gets called for it		|
   |- Now page 1003520 is Clean.	|
   |					| btrfs_setattr()
   |					| |- btrfs_setsize()
   |					|    |- truncate_setsize()
   |					|       New i_size is 18388
   |- __extent_writepage()		|
   |  |- page_offset() > i_size		|
      |- btrfs_invalidatepage()		|
	 |- Page is clean, so no qgroup |
	    callback executed

This means, the qgroup reserved data space is not properly released in
btrfs_invalidatepage() as the page is Clean.

[FIX]
Instead of checking the dirty bit of a page, call
btrfs_qgroup_free_data() unconditionally in btrfs_invalidatepage().

As qgroup rsv are completely bound to the QGROUP_RESERVED bit of
io_tree, not bound to page status, thus we won't cause double freeing
anyway.

Fixes: 0b34c261 ("btrfs: qgroup: Prevent qgroup->reserved from going subzero")
CC: stable@vger.kernel.org # 4.14+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

parent 580c079b

fs/btrfs/inode.c

+10 −13

Original line number	Diff line number	Diff line
		@@ -8136,19 +8136,16 @@ again:
		/*
		* Qgroup reserved space handler
		* Page here will be either
		* 1) Already written to disk
		* In this case, its reserved space is released from data rsv map
		* and will be freed by delayed_ref handler finally.
		* So even we call qgroup_free_data(), it won't decrease reserved
		* space.
		* 2) Not written to disk
		* This means the reserved space should be freed here. However,
		* if a truncate invalidates the page (by clearing PageDirty)
		* and the page is accounted for while allocating extent
		* in btrfs_check_data_free_space() we let delayed_ref to
		* free the entire extent.
		*/
		if (PageDirty(page))
		* 1) Already written to disk or ordered extent already submitted
		* Then its QGROUP_RESERVED bit in io_tree is already cleaned.
		* Qgroup will be handled by its qgroup_record then.
		* btrfs_qgroup_free_data() call will do nothing here.
		*
		* 2) Not written to disk yet
		* Then btrfs_qgroup_free_data() call will clear the QGROUP_RESERVED
		* bit of its io_tree, and free the qgroup reserved data space.
		* Since the IO will never happen for this page.
		*/
		btrfs_qgroup_free_data(inode, NULL, page_start, PAGE_SIZE);
		if (!inode_evicting) {
		clear_extent_bit(tree, page_start, page_end, EXTENT_LOCKED \|

Admin message