CVE-2026-53320: nilfs2: reject zero bd_oblocknr in nilfs_ioctl_mark_blocks_dirty()

In the Linux kernel, the following vulnerability has been resolved:

nilfs2: reject zero bdoblocknr in nilfsioctlmarkblocksdirty()

nilfsioctlmarkblocksdirty() uses bdoblocknr to detect dead blocks by comparing it with the current block number bdblocknr. If they differ, the block is considered dead and skipped.

However, bdoblocknr should never be 0 since block 0 typically stores the primary superblock and is never a valid GC target block. A corrupted ioctl request with bdoblocknr set to 0 causes the comparison to incorrectly match when the lookup returns -ENOENT and sets bdblocknr to 0, bypassing the dead block check and calling nilfsbmapmark() on a non-existent block. This causes nilfsbtreedolookup() to return -ENOENT, triggering the WARNON(ret == -ENOENT).

Fix this by rejecting ioctl requests with bdoblocknr set to 0 at the beginning of each iteration.

[ryusuke: slightly modified the commit message and comments for accuracy]