CVE-2026-72427

01/Description

In the Linux kernel, the following vulnerability has been resolved: bpf: Fix effective prog array index with BPF_F_PREORDER replace_effective_prog() and purge_effective_progs() located the slot in the effective array by walking the program hlist and counting entries linearly. That count does not match the array layout: compute_effective_ progs() places BPF_F_PREORDER programs at the front (ancestor cgroup first, attach order within a cgroup) and the rest after them (descendant cgroup first). So when a preorder program is present, the linear hlist position no longer equals the program's index in the effective array. For replace_effective_prog() (bpf_link_update()) this overwrote the wrong slot, corrupting the effective order. For purge_effective_progs(), it could dummy out a slot belonging to a different program and leave the detached program in the array while bpf_prog_put() drops its reference, i.e. a use-after-free. Fix both by replaying compute_effective_progs()'s placement (including the per-cgroup preorder reversal) in a shared effective_prog_pos() helper. Identify the entry by its struct bpf_prog_list pointer rather than by (prog, link) value, so the lookup resolves to exactly the attachment the syscall selected even when the same bpf_prog is attached to several cgroups in the hierarchy.

02/References5

• git.kernel.org/stable/c/41b4320b84fdafe1ab586b06453d30d50415db59
• git.kernel.org/stable/c/525e408c27ae714e538b8c608c3a974df3ab6c92
• git.kernel.org/stable/c/9697db03e010391c55ae75192cbdf30c5a72c114
• git.kernel.org/stable/c/b584f107ab90222bd825dcb4c5977326ff684109
• git.kernel.org/stable/c/f08aaee3152d0dfc578b3f2586932d82062701dd