Lee sent a patchset to update Futex for 4.9, see https://www.spinics.net/lists/stable/msg443081.html, Then Xiaoming sent a follow-up patch for it, see https://lore.kernel.org/lkml/20210225093120.GD641347@dell/. These patchsets may also resolve following issues in 4.4.260 which have been reported in 4.9, see https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/?h=linux-4.4.y&id=319f66f08de1083c1fe271261665c209009dd65a > /* > * The task is on the way out. When the futex state is > * FUTEX_STATE_DEAD, we know that the task has finished > * the cleanup: > */ > int ret = (p->futex_state = FUTEX_STATE_DEAD) ? -ESRCH : -EAGAIN; Here may be: int ret = (p->futex_state == FUTEX_STATE_DEAD) ? -ESRCH : -EAGAIN; > raw_spin_unlock_irq(&p->pi_lock); > /* > * If the owner task is between FUTEX_STATE_EXITING and > * FUTEX_STATE_DEAD then store the task pointer and keep > * the reference on the task struct. The calling code will > * drop all locks, wait for the task to reach > * FUTEX_STATE_DEAD and then drop the refcount. This is > * required to prevent a live lock when the current task > * preempted the exiting task between the two states. > */ > if (ret == -EBUSY) And here, the variable "ret" may only be "-ESRCH" or "-EAGAIN", but not "-EBUSY". > *exiting = p; > else > put_task_struct(p); Since 074e7d515783 ("futex: Ensure the correct return value from futex_lock_pi()") has been merged in 4.4.260, I send the remain 3 patches. Peter Zijlstra (1): futex: Change locking rules Thomas Gleixner (2): futex: Cure exit race futex: fix dead code in attach_to_pi_owner() kernel/futex.c | 209 +++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 177 insertions(+), 32 deletions(-) -- 2.25.4
