btrfs: reject root items with drop_progress and zero drop_level

[BUG]
When recovering relocation at mount time, merge_reloc_root() and btrfs_drop_snapshot() both use BUG_ON(level == 0) to guard against an impossible state: a non-zero drop_progress combined with a zero drop_level in a root_item, which can be triggered:

------------[ cut here ]------------
kernel BUG at fs/btrfs/relocation.c:1545!
Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI
CPU: 1 UID: 0 PID: 283 ... Tainted: 6.18.0+ #16 PREEMPT(voluntary)
Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
Hardware name: QEMU Ubuntu 24.04 PC v2, BIOS 1.16.3-debian-1.16.3-2
RIP: 0010:merge_reloc_root+0x1266/0x1650 fs/btrfs/relocation.c:1545
Code: ffff0000 00004589 d7e9acfa ffffe8a1 79bafebe 02000000
Call Trace:
  merge_reloc_roots+0x295/0x890 fs/btrfs/relocation.c:1861
  btrfs_recover_relocation+0xd6e/0x11d0 fs/btrfs/relocation.c:4195
  btrfs_start_pre_rw_mount+0xa4d/0x1810 fs/btrfs/disk-io.c:3130
  open_ctree+0x5824/0x5fe0 fs/btrfs/disk-io.c:3640
  btrfs_fill_super fs/btrfs/super.c:987 [inline]
  btrfs_get_tree_super fs/btrfs/super.c:1951 [inline]
  btrfs_get_tree_subvol fs/btrfs/super.c:2094 [inline]
  btrfs_get_tree+0x111c/0x2190 fs/btrfs/super.c:2128
  vfs_get_tree+0x9a/0x370 fs/super.c:1758
  fc_mount fs/namespace.c:1199 [inline]
  do_new_mount_fc fs/namespace.c:3642 [inline]
  do_new_mount fs/namespace.c:3718 [inline]
  path_mount+0x5b8/0x1ea0 fs/namespace.c:4028
  do_mount fs/namespace.c:4041 [inline]
  __do_sys_mount fs/namespace.c:4229 [inline]
  __se_sys_mount fs/namespace.c:4206 [inline]
  __x64_sys_mount+0x282/0x320 fs/namespace.c:4206
  ...
RIP: 0033:0x7f969c9a8fde
Code: 0f1f4000 48c7c2b0 fffffff7 d8648902 b8ffffff ffc3660f
---[ end trace 0000000000000000 ]---

[CAUSE]
A non-zero drop_progress.objectid means an interrupted btrfs_drop_snapshot() left a resume point on disk, and in that case drop_level must be greater than 0 because the checkpoint is only saved at internal node levels.

Although this invariant is enforced when the kernel writes the root item, it is not validated when the root item is read back from disk. That allows on-disk corruption to provide an invalid state with drop_progress.objectid != 0 and drop_level == 0.

When relocation recovery later processes such a root item, merge_reloc_root() reads drop_level and hits BUG_ON(level == 0). The same invalid metadata can also trigger the corresponding BUG_ON() in btrfs_drop_snapshot().

[FIX]
Fix this by validating the root_item invariant in tree-checker when reading root items from disk: if drop_progress.objectid is non-zero, drop_level must also be non-zero. Reject such malformed metadata with -EUCLEAN before it reaches merge_reloc_root() or btrfs_drop_snapshot() and triggers the BUG_ON.

Also fix the related tree-checker error message to report "invalid root drop_level" instead of the misleading "invalid root level".

The bug is reproducible on 7.0.0-rc2-next-20260310 with our dynamic metadata fuzzing tool that corrupts btrfs metadata at runtime. After the fix, the same corruption is correctly rejected by tree-checker and the BUG_ON is no longer triggered.

Fixes 259ee7754b67 ("btrfs: tree-checker: Add ROOT_ITEM check")

Cc [email protected] # 5.3+

Signed-off-by ZhengYuan Huang <[email protected]>

---
Reproduction (v6.18, x86_64, KASAN)
===================================
The PoC is relatively large, so it is provided separately through google drive:
https://drive.google.com/drive/folders/1Rto3DUtjUTOg3bjFeH5G2Kl8Li6OuzsG

To reproduce the issue:
1. Build the ublk helper program from the ublk codebase, which is

	 used to provide the runtime corruption capability:
	  g++ -std=c++20 -fcoroutines -O2 -o standalone_replay \
       standalone_replay_btrfs.cpp targets/ublksrv_tgt.cpp \
       -I. -Iinclude -Itargets/include \
       -L./lib/.libs -lublksrv -luring -lpthread

2. Attach the crafted image through ublk:
./standalone_replay add -t loop -f /path/to/image
3. Mount the image:
mount -o loop /path/to/image /mnt
This reliably reproduces the bug.

---
  fs/btrfs/tree-checker.c | 19 ++++++++++++++++++-
  1 file changed, 18 insertions(+), 1 deletion(-)

fs/btrfs/tree-checker.c +18 -1

index dd274f67ad7f..a8c568b10432 100644
@@ -1256,10 +1256,27 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
  	}
  	if (unlikely(btrfs_root_drop_level(&ri) >= BTRFS_MAX_LEVEL)) {
  		generic_err(leaf, slot,
-			    "invalid root level, have %u expect [0, %u]",
+			    "invalid root drop_level, have %u expect [0, %u]",
  			    btrfs_root_drop_level(&ri), BTRFS_MAX_LEVEL - 1);
  		return -EUCLEAN;
  	}
+	/*
+	 * If drop_progress.objectid is non-zero, a btrfs_drop_snapshot() was
+	 * interrupted and the resume point was recorded in drop_progress and
+	 * drop_level.  In that case drop_level must be >= 1: level 0 is the
+	 * leaf level and drop_snapshot never saves a checkpoint there (it
+	 * only records checkpoints at internal node levels in DROP_REFERENCE
+	 * stage).  A zero drop_level combined with a non-zero drop_progress
+	 * objectid indicates on-disk corruption and would cause a BUG_ON in
+	 * merge_reloc_root() and btrfs_drop_snapshot() at mount time.
+	 */
+	if (unlikely(btrfs_disk_key_objectid(&ri.drop_progress) != 0 &&
+		     btrfs_root_drop_level(&ri) == 0)) {
+		generic_err(leaf, slot,
+			    "invalid root drop_level 0 with non-zero drop_progress objectid %llu",
+			    btrfs_disk_key_objectid(&ri.drop_progress));
+		return -EUCLEAN;
+	}
 
  	/* Flags check */
  	if (unlikely(btrfs_root_flags(&ri) & ~valid_root_flags)) {

Quoted 9 lines

[FIX]
Fix this by validating the root_item invariant in tree-checker when reading root items from disk: if drop_progress.objectid is non-zero, drop_level must also be non-zero. Reject such malformed metadata with -EUCLEAN before it reaches merge_reloc_root() or btrfs_drop_snapshot() and triggers the BUG_ON.

Also fix the related tree-checker error message to report "invalid root drop_level" instead of the misleading "invalid root level".

The only "fix" part I can see is the fix of the message from drop_level.

If you really want to do that, please send out a fix dedicated for that single line.

Otherwise you're adding a new check. Please do not mix fix and new check into one patch.