In the Linux kernel, the following vulnerability has been resolved:
configfs_lookup(): dont leave ->s_dentry dangling on failure
Normally ->s_dentry is cleared when dentry its pointing to becomes negative (on eviction, realistically). However, that only happens if dentry gets to be positive in the first place; in case of inode allocation failure dentry never becomes positive, so ->d_iput() is not called at all.
We do part of what normally wouldve been done by configfs_d_iput() (dropping the reference to configfs_dirent) manually, but we do not clear ->s_dentry there. Sloppy as it is, it does not matter in case of configfs_create_{dir,link}() - there configfs_dirent does not survive dropping the sole reference to it.
However, for configfs_lookup() it does survive, with a dangling pointer to soon to be freed dentry sitting it its ->s_dentry.
Subsequent getdents(2) in that directory will end up dereferencing that pointer in order to pick the inode number. Use after free…
This is the minimal fix; the right approach is to set the linkage between dentry and configfs_dirent only after we know that we have an inode, but that takes more surgery and the bug had been there since 2006, so…
References
- https://git.kernel.org/stable/c/10da12d352b7b2bb330a8609fdda9a58bf0e9856
- https://git.kernel.org/stable/c/3e83b2203aa59bd279e4f677ec793d49dc9d019e
- https://git.kernel.org/stable/c/57088b06109f3222963c639d8d743f42c2899b13
- https://git.kernel.org/stable/c/9c747dcee164ead300de90550ad9e4122f0d1bbb
- https://git.kernel.org/stable/c/b6e9c82522ddaa3ac0706b295ff4a71975d4f883
- https://git.kernel.org/stable/c/c3b073a209a9baa691b744318ac929fecdd8847c
- https://git.kernel.org/stable/c/eee07d769da5ac4e4f7bd0bc17828646a318d499