On Thu, Nov 29, 2018 at 11:17:14PM +0100, Peter Zijlstra wrote:
> On Thu, Nov 29, 2018 at 01:34:21PM -0800, Davidlohr Bueso wrote:
> > I messed up something such that waiman was not in the thread. Ccing.
> >
> > > On Thu, 29 Nov 2018, Waiman Long wrote:
> > >
> > > > That can be costly for x86 which will now have 2 locked instructions.
> > >
> > > Yeah, and when used as an actual queue we should really start to notice.
> > > Some users just have a single task in the wake_q because avoiding the cost
> > > of wake_up_process() with locks held is significant.
> > >
> > > How about instead of adding the barrier before the cmpxchg, we do it
> > > in the failed branch, right before we return. This is the uncommon
> > > path.
> > >
> > > Thanks,
> > > Davidlohr
> > >
> > > diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> > > index 091e089063be..0d844a18a9dc 100644
> > > --- a/kernel/sched/core.c
> > > +++ b/kernel/sched/core.c
> > > @@ -408,8 +408,14 @@ void wake_q_add(struct wake_q_head *head, struct
> > > task_struct *task)
> > > * This cmpxchg() executes a full barrier, which pairs with the full
> > > * barrier executed by the wakeup in wake_up_q().
> > > */
> > > - if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL))
> > > + if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL)) {
> > > + /*
> > > + * Ensure, that when the cmpxchg() fails, the corresponding
> > > + * wake_up_q() will observe our prior state.
> > > + */
> > > + smp_mb__after_atomic();
> > > return;
> > > + }
>
> So wake_up_q() does:
>
> wake_up_q():
> node->next = NULL;
> /* implied smp_mb */
> wake_up_process();
>
> So per the cross your variables 'rule', this side then should do:
>
> wake_q_add():
> /* wake_cond = true */
> smp_mb()
> cmpxchg_relaxed(&node->next, ...);
>
> So that the ordering pivots around node->next.
>
> Either we see NULL and win the cmpxchg (in which case we'll do the
> wakeup later) or, when we fail the cmpxchg, we must observe what came
> before the failure.
>
> If it wasn't so damn late, I'd try and write a litmus test for this,
> because now I'm starting to get confused -- also probably because it's
> late.
The above description suggests:
C wake_up_q-wake_q_add
{
int next = 0;
int y = 0;
}
P0(int *next, int *y)
{
int r0;
/* in wake_up_q() */
WRITE_ONCE(*next, 1); /* node->next = NULL */
smp_mb(); /* implied by wake_up_process() */
r0 = READ_ONCE(*y);
}
P1(int *next, int *y)
{
int r1;
/* in wake_q_add() */
WRITE_ONCE(*y, 1); /* wake_cond = true */
smp_mb__before_atomic();
r1 = cmpxchg_relaxed(next, 1, 2);
}
exists (0:r0=0 /\ 1:r1=0)
This "exists" clause cannot be satisfied according to the LKMM:
Test wake_up_q-wake_q_add Allowed
States 3
0:r0=0; 1:r1=1;
0:r0=1; 1:r1=0;
0:r0=1; 1:r1=1;
No
Witnesses
Positive: 0 Negative: 3
Condition exists (0:r0=0 /\ 1:r1=0)
Observation wake_up_q-wake_q_add Never 0 3
Time wake_up_q-wake_q_add 0.00
Hash=72d85545f97ef7fd35c8928259225ee0
(TBH, I'm not sure what "y" (you denoted it "wake_cond") is pointing to
here/is modeling, but I might have missed some previous remarks...)
Andrea
>
> In any case, I think you patch is 'wrong' because it puts the barrier on
> the wrong side of the cmpxchg() (after, as opposed to before).
