CVE-2026-98984

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

dcache: don't put a mountpoint on a dentry that's being removed

rmdir(), unlink() and rename() call dont mount() on the victim and then detach mounts() with the victim's inode locked. do lock mount() takes the inode lock of the mountpoint and checks cant mount() so a mount can't show up after detach mounts().

But attach recursive mnt() makes a second mountpoint for the root of the source mount so that the mounts already located at the destination can be put on top of it. No inode is locked for that one and d set mounted() only refuses a dentry that is unlinked. Between detach mounts() and d delete() the victim is still hashed:

rmrace: b passed dont mount() and detach mounts(), sleeping T2: move mount(S, /tmp/plcant/x, BENEATH) = 0 errno 0 () T1: rmdir(/tmp/plcant/d/b) = 0 errno 0 109 107 0:61 /d/b//deleted /tmp/plcant/x rw,relatime - tmpfs tmpfs rw 108 109 0:63 / /tmp/plcant/x rw,relatime - tmpfs T rw

A bind mount of the directory that's being removed is moved beneath an existing mount while the rmdir() is located between the two calls. The mount ends up on the removed directory and nothing will ever detach it.

Check cant mount() in d set mounted() as well. dont mount() raises the flag under d lock before detach mounts() runs so either the mountpoint is set first and detach mounts() finds the mount or the flag is seen and the mount is refused.