CVE-2026-80775: futex: Fix race on the initial mm->futex.phash.ref allocation
In the Linux kernel, the following vulnerability has been resolved:
futex: Fix race on the initial mm->futex.phash.ref allocation
futexhashallocate() allocates mm->futex.phash.ref without any locking. Commit d9b05321e21e ("futex: Move futexhashfree() back to mmput()") moved the allocation here and assumed that the process has just a single thread at this point.
Commit ee9dce44362b ("futex: Drop CLONETHREAD requirement for private default hash alloc") widened needfutexhashallocatedefault() to cover any CLONEVM clone, but left out vfork because the parent is suspended and cannot race.
That no longer holds once vfork is nested. If a vfork child calls vfork again and is then killed with SIGKILL, the parent is released from its vfork wait and runs concurrently with the grandchild in the same mm. Neither of them went through futexhashallocatedefault().
When both call prctl(PRFUTEXHASH, PRFUTEXHASHSETSLOTS) at the same time, each one sees mm->futex.phash.ref as NULL and stores its own percpu counter. Only the last store survives. The counter stored first is no longer reachable from the mm, so the references on it are not seen by futexrefatomicend(). A private hash that still has references is then considered dead and freed, and a task that still holds one of its buckets writes into freed memory in futexqlock().
Store the counter once with cmpxchg() and let the loser freepercpu() its own. The initial reference has to be taken before the store, otherwise another task can install a private hash while the counter is still 0.
Affected Software
Event History
Frequently Asked Questions
What conditions are required to trigger the race?
The scenario requires a nested vfork: a vfork child creates another vfork child and is killed with SIGKILL, allowing the original parent to run concurrently with the grandchild in the same memory map. Both processes must then concurrently call prctl(PR_FUTEX_HASH, PR_FUTEX_HASH_SET_SLOTS).
Are ordinary multithreaded processes sufficient to expose this issue?
The described race specifically depends on nested vfork behavior and concurrent execution in a shared mm after the intermediate vfork child is killed. The description does not indicate that ordinary threading alone triggers this allocation race.