locking/futex: Allow low-level atomic operations to return -EAGAIN
Some futex() operations, including FUTEX_WAKE_OP, require the kernel to perform an atomic read-modify-write of the futex word via the userspace mapping. These operations are implemented by each architecture in arch_futex_atomic_op_inuser() and futex_atomic_cmpxchg_inatomic(), which are called in atomic context with the relevant hash bucket locks held. Although these routines may return -EFAULT in response to a page fault generated when accessing userspace, they are expected to succeed (i.e. return 0) in all other cases. This poses a problem for architectures that do not provide bounded forward progress guarantees or fairness of contended atomic operations and can lead to starvation in some cases. In these problematic scenarios, we must return back to the core futex code so that we can drop the hash bucket locks and reschedule if necessary, much like we do in the case of a page fault. Allow architectures to return -EAGAIN from their implementations of arch_futex_atomic_op_inuser() and futex_atomic_cmpxchg_inatomic(), which will cause the core futex code to reschedule if necessary and return back to the architecture code later on. Cc: <stable@kernel.org> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
This commit is contained in:
parent
84ff7a09c3
commit
6b4f4bc9cb
120
kernel/futex.c
120
kernel/futex.c
|
@ -1311,13 +1311,15 @@ static int lookup_pi_state(u32 __user *uaddr, u32 uval,
|
|||
|
||||
static int lock_pi_update_atomic(u32 __user *uaddr, u32 uval, u32 newval)
|
||||
{
|
||||
int err;
|
||||
u32 uninitialized_var(curval);
|
||||
|
||||
if (unlikely(should_fail_futex(true)))
|
||||
return -EFAULT;
|
||||
|
||||
if (unlikely(cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)))
|
||||
return -EFAULT;
|
||||
err = cmpxchg_futex_value_locked(&curval, uaddr, uval, newval);
|
||||
if (unlikely(err))
|
||||
return err;
|
||||
|
||||
/* If user space value changed, let the caller retry */
|
||||
return curval != uval ? -EAGAIN : 0;
|
||||
|
@ -1502,10 +1504,8 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_
|
|||
if (unlikely(should_fail_futex(true)))
|
||||
ret = -EFAULT;
|
||||
|
||||
if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) {
|
||||
ret = -EFAULT;
|
||||
|
||||
} else if (curval != uval) {
|
||||
ret = cmpxchg_futex_value_locked(&curval, uaddr, uval, newval);
|
||||
if (!ret && (curval != uval)) {
|
||||
/*
|
||||
* If a unconditional UNLOCK_PI operation (user space did not
|
||||
* try the TID->0 transition) raced with a waiter setting the
|
||||
|
@ -1700,32 +1700,32 @@ retry_private:
|
|||
double_lock_hb(hb1, hb2);
|
||||
op_ret = futex_atomic_op_inuser(op, uaddr2);
|
||||
if (unlikely(op_ret < 0)) {
|
||||
|
||||
double_unlock_hb(hb1, hb2);
|
||||
|
||||
#ifndef CONFIG_MMU
|
||||
if (!IS_ENABLED(CONFIG_MMU) ||
|
||||
unlikely(op_ret != -EFAULT && op_ret != -EAGAIN)) {
|
||||
/*
|
||||
* we don't get EFAULT from MMU faults if we don't have an MMU,
|
||||
* but we might get them from range checking
|
||||
* we don't get EFAULT from MMU faults if we don't have
|
||||
* an MMU, but we might get them from range checking
|
||||
*/
|
||||
ret = op_ret;
|
||||
goto out_put_keys;
|
||||
#endif
|
||||
|
||||
if (unlikely(op_ret != -EFAULT)) {
|
||||
ret = op_ret;
|
||||
goto out_put_keys;
|
||||
}
|
||||
|
||||
if (op_ret == -EFAULT) {
|
||||
ret = fault_in_user_writeable(uaddr2);
|
||||
if (ret)
|
||||
goto out_put_keys;
|
||||
}
|
||||
|
||||
if (!(flags & FLAGS_SHARED))
|
||||
if (!(flags & FLAGS_SHARED)) {
|
||||
cond_resched();
|
||||
goto retry_private;
|
||||
}
|
||||
|
||||
put_futex_key(&key2);
|
||||
put_futex_key(&key1);
|
||||
cond_resched();
|
||||
goto retry;
|
||||
}
|
||||
|
||||
|
@ -2350,7 +2350,7 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
|
|||
u32 uval, uninitialized_var(curval), newval;
|
||||
struct task_struct *oldowner, *newowner;
|
||||
u32 newtid;
|
||||
int ret;
|
||||
int ret, err = 0;
|
||||
|
||||
lockdep_assert_held(q->lock_ptr);
|
||||
|
||||
|
@ -2421,14 +2421,17 @@ retry:
|
|||
if (!pi_state->owner)
|
||||
newtid |= FUTEX_OWNER_DIED;
|
||||
|
||||
if (get_futex_value_locked(&uval, uaddr))
|
||||
goto handle_fault;
|
||||
err = get_futex_value_locked(&uval, uaddr);
|
||||
if (err)
|
||||
goto handle_err;
|
||||
|
||||
for (;;) {
|
||||
newval = (uval & FUTEX_OWNER_DIED) | newtid;
|
||||
|
||||
if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval))
|
||||
goto handle_fault;
|
||||
err = cmpxchg_futex_value_locked(&curval, uaddr, uval, newval);
|
||||
if (err)
|
||||
goto handle_err;
|
||||
|
||||
if (curval == uval)
|
||||
break;
|
||||
uval = curval;
|
||||
|
@ -2456,23 +2459,37 @@ retry:
|
|||
return 0;
|
||||
|
||||
/*
|
||||
* To handle the page fault we need to drop the locks here. That gives
|
||||
* the other task (either the highest priority waiter itself or the
|
||||
* task which stole the rtmutex) the chance to try the fixup of the
|
||||
* pi_state. So once we are back from handling the fault we need to
|
||||
* check the pi_state after reacquiring the locks and before trying to
|
||||
* do another fixup. When the fixup has been done already we simply
|
||||
* return.
|
||||
* In order to reschedule or handle a page fault, we need to drop the
|
||||
* locks here. In the case of a fault, this gives the other task
|
||||
* (either the highest priority waiter itself or the task which stole
|
||||
* the rtmutex) the chance to try the fixup of the pi_state. So once we
|
||||
* are back from handling the fault we need to check the pi_state after
|
||||
* reacquiring the locks and before trying to do another fixup. When
|
||||
* the fixup has been done already we simply return.
|
||||
*
|
||||
* Note: we hold both hb->lock and pi_mutex->wait_lock. We can safely
|
||||
* drop hb->lock since the caller owns the hb -> futex_q relation.
|
||||
* Dropping the pi_mutex->wait_lock requires the state revalidate.
|
||||
*/
|
||||
handle_fault:
|
||||
handle_err:
|
||||
raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
|
||||
spin_unlock(q->lock_ptr);
|
||||
|
||||
switch (err) {
|
||||
case -EFAULT:
|
||||
ret = fault_in_user_writeable(uaddr);
|
||||
break;
|
||||
|
||||
case -EAGAIN:
|
||||
cond_resched();
|
||||
ret = 0;
|
||||
break;
|
||||
|
||||
default:
|
||||
WARN_ON_ONCE(1);
|
||||
ret = err;
|
||||
break;
|
||||
}
|
||||
|
||||
spin_lock(q->lock_ptr);
|
||||
raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
|
||||
|
@ -3041,10 +3058,8 @@ retry:
|
|||
* A unconditional UNLOCK_PI op raced against a waiter
|
||||
* setting the FUTEX_WAITERS bit. Try again.
|
||||
*/
|
||||
if (ret == -EAGAIN) {
|
||||
put_futex_key(&key);
|
||||
goto retry;
|
||||
}
|
||||
if (ret == -EAGAIN)
|
||||
goto pi_retry;
|
||||
/*
|
||||
* wake_futex_pi has detected invalid state. Tell user
|
||||
* space.
|
||||
|
@ -3059,9 +3074,19 @@ retry:
|
|||
* preserve the WAITERS bit not the OWNER_DIED one. We are the
|
||||
* owner.
|
||||
*/
|
||||
if (cmpxchg_futex_value_locked(&curval, uaddr, uval, 0)) {
|
||||
if ((ret = cmpxchg_futex_value_locked(&curval, uaddr, uval, 0))) {
|
||||
spin_unlock(&hb->lock);
|
||||
switch (ret) {
|
||||
case -EFAULT:
|
||||
goto pi_faulted;
|
||||
|
||||
case -EAGAIN:
|
||||
goto pi_retry;
|
||||
|
||||
default:
|
||||
WARN_ON_ONCE(1);
|
||||
goto out_putkey;
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
|
@ -3075,6 +3100,11 @@ out_putkey:
|
|||
put_futex_key(&key);
|
||||
return ret;
|
||||
|
||||
pi_retry:
|
||||
put_futex_key(&key);
|
||||
cond_resched();
|
||||
goto retry;
|
||||
|
||||
pi_faulted:
|
||||
put_futex_key(&key);
|
||||
|
||||
|
@ -3435,6 +3465,7 @@ err_unlock:
|
|||
static int handle_futex_death(u32 __user *uaddr, struct task_struct *curr, int pi)
|
||||
{
|
||||
u32 uval, uninitialized_var(nval), mval;
|
||||
int err;
|
||||
|
||||
/* Futex address must be 32bit aligned */
|
||||
if ((((unsigned long)uaddr) % sizeof(*uaddr)) != 0)
|
||||
|
@ -3444,7 +3475,9 @@ retry:
|
|||
if (get_user(uval, uaddr))
|
||||
return -1;
|
||||
|
||||
if ((uval & FUTEX_TID_MASK) == task_pid_vnr(curr)) {
|
||||
if ((uval & FUTEX_TID_MASK) != task_pid_vnr(curr))
|
||||
return 0;
|
||||
|
||||
/*
|
||||
* Ok, this dying thread is truly holding a futex
|
||||
* of interest. Set the OWNER_DIED bit atomically
|
||||
|
@ -3456,6 +3489,7 @@ retry:
|
|||
* userspace.
|
||||
*/
|
||||
mval = (uval & FUTEX_WAITERS) | FUTEX_OWNER_DIED;
|
||||
|
||||
/*
|
||||
* We are not holding a lock here, but we want to have
|
||||
* the pagefault_disable/enable() protection because
|
||||
|
@ -3465,11 +3499,23 @@ retry:
|
|||
* does not guarantee R/W access. If that fails we
|
||||
* give up and leave the futex locked.
|
||||
*/
|
||||
if (cmpxchg_futex_value_locked(&nval, uaddr, uval, mval)) {
|
||||
if ((err = cmpxchg_futex_value_locked(&nval, uaddr, uval, mval))) {
|
||||
switch (err) {
|
||||
case -EFAULT:
|
||||
if (fault_in_user_writeable(uaddr))
|
||||
return -1;
|
||||
goto retry;
|
||||
|
||||
case -EAGAIN:
|
||||
cond_resched();
|
||||
goto retry;
|
||||
|
||||
default:
|
||||
WARN_ON_ONCE(1);
|
||||
return err;
|
||||
}
|
||||
}
|
||||
|
||||
if (nval != uval)
|
||||
goto retry;
|
||||
|
||||
|
@ -3479,7 +3525,7 @@ retry:
|
|||
*/
|
||||
if (!pi && (uval & FUTEX_WAITERS))
|
||||
futex_wake(uaddr, 1, 1, FUTEX_BITSET_MATCH_ANY);
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
|
Loading…
Reference in New Issue